Datasets:

ianiket23

/

podcast_llama_chat_format-5k

like 0

<s>[INST] So you're talking about any kind of recommended system, let's look at the Youtube algorithm, facebook, anything that recommends content you should watch next and it's fascinating to think that there's some aspects of human behavior that you can, you know, say a problem of is this person hold the republican beliefs or democratic beliefs and it's just trivial, uh, that's an objective function and you can optimize and you can measure and you can turn everybody into republican, [/INST] everybody into a bit of it's true. So the human mind is very, if you look at the human mind is a kind of computer program, it is a very large exploits. Face right has many, many vulnerabilities. Whereas you can control it. For instance, when it comes to your political beliefs, this is very much tied to your identity. So for instance, if I'm in control of your news feed on your favorite social media platforms, uh, this is actually where you're getting your news from. And I can of course I can, I can choose to only show you news that will make you see the world in a specific wear right. But I can also um, uh, you know, create incentives for you to, to to post about some political beliefs. And then when I, when I get you to express a statement, if it's a statement that me as a, as a controller, I want, I want to reinforce, I can just show it to people will agree and they will like it and that will reinforce the statement in your mind. If this is a statement, I want you to uh, this belief. I I want you to abandon, I can on the other hand, show it to uh, openness wright will attack you and because they attack you at the very least, next time you will think twice about posting it, but maybe you will even uh you know, start believing this because you got to push back, right? So there are many ways in which uh social media platforms can potentially control your opinions and today. Uh the so all of these things are already being controlled by ai algorithms, algorithms do not have any explicit political goal today. Well, potentially the kids like uh uh if some totalitarian government takes over, you know, social media platforms and decides that, you know, now we're going to use this knowledge is from as civilians, but also for my opinion controlling behavior control, very bad things could happen. Um but what's really uh fascinating and actually quite concerning is that even with that an explicit intent to manipulate you're already seeing very dangerous dynamics in terms of has this content recommendation algorithms behave because right now, them the goal, the objective function of these algorithms is to maximize engagement, right? Which seems fairly uh innocuous at first. Right. However, uh it is not because content that will maximally engage people, you know, get people to react in an emotional way to get people to click on something. It is very often um content that, you know, it's not healthy to the public discourse for instance, um fake news are far more likely to get you to click on them than real news simply because they are not um constrained to reality. So they can be as outrageous as as a a surprising as as good stories as you want because the artificial right? Yeah. To </s><s>[INST] me, that's an exciting world because so much good can come. So there's an opportunity to educate people. You can uh balance people's worldview with other ideas. So there's so many objective functions. The space of objective functions that create better civilizations is large, arguably infinite. But there's also a large space that creates division and uh and uh and destruction civil war, a lot of bad stuff. And the worry is uh naturally probably that space is bigger, first of all, and if we don't explicitly think about what kind of effects are going to be observed from different objective functions, then we're gonna get into trouble. But the question is, how do we how do we get into rooms and have discussions so inside google inside facebook, inside twitter and think about, okay, how can we drive up engagement and at the same time create a good [/INST] society, </s><s>[INST] is it even possible to have that kind of philosophical discussion? [/INST] I think you can try. So from my perspective, I would feel rather uncomfortable with companies that are in control of these uh news feed algorithms, with them making explicit decisions to manipulate people's opinions are behaviors, even if the intent is good because that's that's a very totalitarian mindset. So instead, what I would like to see, it's probably never going to happen because it's not very realistic. But that's actually something I really care about. I would like all these algorithms uh to present configuration settings to their users so that their users can actually make the decision about how they want to be impacted by these uh information recommendation content recommendation algorithms. For instance, as a as a user of something like Youtube twitter, maybe I want to maximize learning about a specific topic. Right? So I want the algorithm um to uh feed my curiosity right? Which is in itself a very interesting problem. So instead of maximizing my engagement, it will maximize how fast and how much I'm learning and it will also take into account the accuracy hopefully, you know, of the information I'm learning. So yeah, the user should be able to determine exactly has its algorithms affecting their lives. I I don't want actually any entity making decisions about in which direction they're gonna try to manipulate me. Right? I want I want technology, so ai algorithms are increasingly going to be our interface to a world that is increasingly made of information and I want I want everyone to be in control of this interface to interface with the world on their own terms. So if someone wants as these algorithms to serve, you know, their own personal growth goals. They should be able to configure this algorithm since the show where </s>

<s>[INST] What about so that that's a straightforward criticism of oversimplified models. What about the idea that we innovate our way out of it. So there's totally new. What is there? There's a silly poor example at this time perhaps. But lab grown meat sort of engineered food so completely shifting source of food for civilization. So therefore alleviating some of the pressure on agriculture [/INST] that comes down to the difference that Ellen makes between between producing a prototype and mass producing the private prototype, you can develop the idea very rapidly to put that into production on the scale that's necessary to replace what we're currently doing years and we haven't got years, we've got we might have decades. We certainly haven't got centuries. So in the time frame we've got, I can't see that engineering going from prototype to production levels to replace what we're currently doing in the stable environment, currently destroying. </s><s>[INST] What do you think about the sort of the catastrophic predictions that people that have thought have written about climate have made in the past that I haven't come to [/INST] mainly, unfortunately, involving paul, ehrlich and the population bomb and the predictions paul was made a </s><s>[INST] few individuals or the one individual in that case. So I'm mostly playing devil's advocate conversation, enjoying doing so. Um, I do think uh, I'm in agreement with the majority of the scientific community, but [/INST] you still see that argument made. </s><s>[INST] I still see the argument made and I also am a little bit worried about the arrogance and the ineffectiveness of the arrogance. [/INST] This is the problem. It's ineffective and that's what worries me because it's all been put into the sea level rise, temperature changes. It's not put into the fragility of the system in which we currently live and the earth will survive. And there's a wonderful science fiction book called The Earth Abides about a world in which humans get wiped out. There's only a tiny band left and then the earth reasserts itself. So the Earth is going to survive us, Will we survive what we do to the earth. That's the question. And my feeling is that we have underplayed the extent to which the civilizations were built have depended upon a relatively stable climate. And then there's that turning point in the maximum in the global average temperature that we evolved right on the top of it. And if we had done nothing, we could find that heading back down towards another ice age could equally destroy the possibility of sedentary life. Example, if we'd never developed fossil fuel based industries, we'd never build super phosphate. So our population would never have reached one billion people and we were still living like a fairly sophisticated animals, but like 17th century level of load on the planet, then we would have gone down that decline and the approaching ice age would have started to wipe out our farming areas, the glaciers would have encroached and we would have been driven out of an agricultural sedentary civilization by that change. So it's just the fact that we evolved on this stable period in the overall temperature cycle of the planet. And that stability is something which just reflects this a turning point in the regular cycle of milosevic cycle, I think it's called. I've forgotten the actual name, but it's a cycle caused by changing the Earth's orbit around the sun reflectivity and so on. That cycle, it's just that tiny top bit that we evolved in. So what we should have done. So that's really useful for us. We should stay at that level now. If we hadn't done it, we'd go back down here and that would be the end of our civilization by an ice age. Instead, we're going up here really rapidly. We're causing a change in temperature compared to the long term cycle 100,000 times faster. So </s><s>[INST] yeah, I mean my my biggest worry is um even subtle changes in climate might result pressure in geopolitical pressures that then lead to nuclear [/INST] war. And that's, I mean, there's an argument that's actually behind, to some extent not the Ukraine war so much, but the arab spring, the wars in Syria, which partially has led to what's happening in Ukraine. So, </s><s>[INST] and our weapons are getting more and more powerful and more and more destructive, more and more nations are having these distinctive weapons and now we're entering cyber space where it's even easier to be [/INST] destructive hyperbaric weapons, which didn't exist in the Second World War. So you don't need nuclear weapons to have catastrophic attacks on each other. So yeah, it's incredibly scary that the warlike side of human nature could be extremely enhanced by climate breakdown. So </s><s>[INST] in this world, on a happy note, uh I don't know how we went from Marxism and stalinism to ecology, but all those are beautiful, complex systems. Uh what is the best form of government would you say on the pub? We talked about the economics of things you ran for office. So you care about politics too. How can politics, what political systems can help us here? [/INST] I think we, first of all have to appreciate where one species on the planet out of millions. And as the intelligence species, we should be enabling harmonious life for those other species as well. </s><s>[INST] Can we actually linger on that? What is you mentioned that we need to acknowledge the value of life on earth? You know? Can we integrate uh the labor theory of value? Can we integrate into that? The value of [/INST] life? Um </s><s>[INST] So there's human [/INST] life, I think if you if you take like that, that structure that I talked about of Marx's use value exchange value, dialectic, the foreground background that only exists. That only works because we're exploiting the free energy. We find in the universe there could be no production system without free energy, which is a, you know, the first law of thermodynamics. Free </s>

<s>[INST] several thousands of them. I don't know. Is there, Could you pull up? There's a video I saw. Yeah, like right here. So can you pause for a second? Uh So so there's march of the penguins. So you always see like penguins huddling together against I mean sorry if I see just metaphors and everything but uh them huddling together against the harshness of the conditions around them. That's very kind of that's like a metaphor for life. Like finding this connection. That's kind of what love is. It's like allows you to forget whatever the absurdity, whatever the suffering of life is together, you get to like huddle for warmth. And that's why I loved the sort of just the the honesty and the intensity of the way penguins just in the middle of like the cold do this and then this video I saw a lonely this is this is misinformation. So the name of the videos lonely deranged penguin, I don't know if he's deranged. So if you play it. So he left his pack and uh and then there's a nice like voice over and you don't need to play it. But he, for some reason left the pack and journeyed out into the mountains. And so the narrator says that he's deranged, he's lost his mind. Now I'd like to project the idea that he's actually, there's so many stories you could think of. He's returning to his homeland. He's an outsider thinking, journey out into the unknown, thinking. He may be able to discover something greater than the tribe. He might be looking for lost love. Why is he deranged immediately? Why has he lost his mind anyway. But this people should look up this video because to me, I might be the only one who romanticizes this. But it's such a nice kind of, it's both a picture of perhaps a mental disorder, which is what the video kind of describes and maybe some deeper explanation. That's not that that has to do with the motivation of uh, of a mind. But yeah, I don't know if I don't know if you have a deeper analysis on this penguin. Well, [/INST] I like you </s><s>[INST] a psychiatrist. [/INST] I would I would want to sit down with the penguin and and and and go through. I want to see the notes from his prior therapist. And but this this actually is, is relevant not knowing what was that penguins motivation we have very clear situations where there are both within an individual. We go through periods of time when we stay in one place and we reap the benefits from what we've built and then we go through periods of foraging, of of of wandering. Even if there may be resources where we are, we have periods of time in our lives where we wander where we where we go in an exploratory mode and different people express that trade in different ways. This is not a human specific trait. If you go down to to the tiny little nematode worm C. Elegans with 302 nervous system cells, </s><s>[INST] they [/INST] go through these phases of foraging and rest and different individuals have different propensity to forage or to rest and stay in one place at the level of the species. That's really good that there's that diversity in their willingness to forge some stay where they are. The species is somewhat on a firm footing then, but some carry a burden a risk for themselves. But it's good for the species that they're explorers and they will venture out the migration patterns that different species blunder into and that turned out to be really good. They weren't logically derived. They most certainly started from something like this. An exploration. </s><s>[INST] Humans do this to you think, [/INST] and we do it too. In fact, it's something we do extremely well. </s><s>[INST] Let's talk about psychiatry a little bit. So in my book, you're a rock star. First of all, for people who don't know your aside from, sort of, um, the neurological view of the brain and neuroscience view of the brain, you're also one of the great psychiatrists of our time. I've always not always, but when I was younger, I dreamed about being a psychiatrist. Um, so it's like, it's like getting to meet your heroes and also, um, you know, getting to meet the people who uh, the best at the top of the world, at the thing you've failed to pursue. So there, so there's, I'm getting a free therapy session on top of that. Okay, so what big picture, what is the practice, the goal, the hope of modern psychiatry? If you could try to describe the discipline As you see it maybe historically, throughout the 20th century, Uh, in contrasting to what it is today, or maybe if you want to describe to what you hope psychiatry becomes, or longs to become in the 21st century. [/INST] It's been an interesting journey. Uh, psychiatry started out pretty firmly grounded in neurology and pathology. Some of the initial uh, founders, effectively of the field were very well grounded in microscopy, looking at cells, uh, working with patients, particularly the neurological side, and that's certainly included, you know, Freud and some of his contemporaries, and but they rapidly discovered that what they could work with at the level of cells and microscopy was so far from the realm of what they could get from a human being and what they were getting from the human being was so much more interesting and had was so mysterious and so unknown that many of them just said we're going to inhabit this domain and we're gonna work with the people with their words and understand what we can based on verbal communication because that was the only tool that people really had and that was a very important step for the field. I would say one of the interesting things that came from the early decades of psychiatry really was this distinction between the conscious and the unconscious mind and paying particular attention to the unconscious mind is something that was worthy of consideration that that might be important in explaining people's actions and that. Perhaps even insight into that was valuable in its own right. And out of that psychoanalysis uh became a practice that it was not always focused on cures or treatments but was more focused on insight. What does it, what does it mean? How can we help people understand why they're feeling something or thinking something or dreaming something. And that insight separate even from treatment was was an interesting thing, as long as one was honest about that and said, you know, we're going for understanding, we're going for insight </s>

<s>[INST] It's a hack that that seemed to work 2, 2 in a small way extent out of the box. [/INST] But so small block side was like happy with these very low levels of block size. And the big block side wanted to expand to, you know, at the very least go to four megabytes, then, you know, maybe go maybe 8 20 there's disagreement. So within there as well, um I definitely was favoring the big side um the whole way through as you can probably tell. Um but even though so the argument </s><s>[INST] against the big is that uh it uh makes things more centralized. [/INST] Yes, because fewer people can run and know that verifies the chain. And also because any of these things would require a hard for count. Hard forks are inherently risky. Do you </s><s>[INST] think there's truth to that? [/INST] Um I'm pro hard fork. I think hard forks are actually like in a political economic sense, they're better than soft forks. Well, let's okay, </s><s>[INST] okay. I think it's a beautiful [/INST] principle </s><s>[INST] as stated that soft forks maybe more course coercive than hard forks. This this is not just about Cryptocurrency. This is about [/INST] politics </s><s>[INST] in life. That's fascinating. So you're okay with hard forks. In fact you think hard forks is the right way to make changes because then everybody is forced to make a decision. Do you accept this change or not as opposed to ideas being sneaked in behind the door and your and the decision is forced on [/INST] you. </s><s>[INST] Okay. So, but you know, hard fork, some people say this is when they talk about sort of ethereum is there's some aspect of a hard fork where you're trying to upgrade, uh, what's the airplane while it's flying? And uh, [/INST] I think self works are also upgrading an airplane while it's flying, </s><s>[INST] but it's a smaller upgrades. [/INST] That's yeah, like there's some truth to that. Like there's definitely, there's definitely a bit more risk of like a split as a result of a hard fork than as a result of a soft work. </s><s>[INST] Um, and this split is highly undesirable, right? [/INST] Well, it depends like if it's a split because of a bug, then that's horrible. If it's a split as a result of political differences, then I think like a split is better than, you know, one side being forced to basically just like suck it up and accept the majority position even if it really hates it. </s><s>[INST] Well, there's also political connections throughout the history of the United States. It's like sometimes groups of people that strongly disagree with each other should be forced to work it out, even if they, even when a split seems like an easy thing in the short term. [/INST] It depends. And I think, well for block chains in particular, the costs of people being able to like peaceful, easier go off and do their own thing or much lower right. Like, you know, okay, if you have a country and you have two groups than uh, like often enough, like fighting out the new rules requires, you know, a civil war requires everyone to move and so forth. But on, on a Blockchain, like, you know, the costs are lower. And so, </s><s>[INST] so if you were to look at the way things worked out with the block size wars and there was a split, what is the Bitcoin cash and yeah, would you like you looking putting on your historian, you mentioned offline you like dan Carlin. So if dan Carlin, what to do an episode on the the block size wars, uh, do you think it could have turned out better or do you um, are you okay with the way it turned out? [/INST] I'm definitely disappointed with what happened with the block with the big block side. Um, I think the source of my disappointment is that like one of the things that you notice when just looking at like this political disagreements generally, especially when you have environments where, you know, they're authoritarian, single party dominated and then there's some opposition party and the opposition often has like very legitimate grievances. But at the same time, the thing you notice is that often enough, the opposition just sucks right? Like it just doesn't have, you know, political capacity. It doesn't have like the ability to come up with policy because its entire culture is like designed around resisting much more and then it's designed around like you know, actually debating serious policy trade offs and I worry or I guess not so much worried because it's already happened, I unfortunately think that Bitcoin cash ended up being a victim of this, right? Like first, you know, there was a split with Bitcoin cash and then of course Craig Wright came in and you know, craig wright, was this basically scammer who just keeps on pretending that he is Satoshi Nakamoto, the inventor of Bitcoin, Hey craig Wright's legal team, Do you hear me? Yes, I still think your client is a scammer, so sue me, this is definitely gonna be </s>

<s>[INST] By the way, cal Newport's author of Deep Work, a book I highly recommend will eventually be on this podcast. I talked to him often. He's a constant inspiration. He has his own podcast two on productivity called Deep questions. They should definitely check out. Anyway, each piece of neuro gom is about 1/2 cup of coffee is worth of caffeine and [/INST] you may know that </s><s>[INST] I love caffeine. I also just love coffee and tea makes me feel like home somehow. Anyway, Bureau is offering 50% off when you go to get in your.com and use code lex at checkout. Once again, that's getting euro dot com and use code lex. This show is also sponsored by eight sleep and its pod pro mattress. You can check out at a sleep dot com slash lex to get $200 off it controls temperature with an app And it's packed with sensors and can cool down to as low as 35° on each side of the bed separately anecdotally. It's been a game changer for me. I'm generally the kind of person that's not about material possessions. I don't have many fancy things in my life. So this bed has been a kind of exciting edition into the mix. There's something about the combination of a cool bed surface with uh warm blanket after a long day of focused work that it's just something I look forward to sleep in general for me lately. I don't know if it's related to the bed or not has been full of dreams. So I've been uh exploring the universe, not constrained by the physics of reality quite a bit. Maybe it's the bed. You should try it out and see. I should mention that they can track a bunch of metrics like heart rate variability, but cooling alone honestly is worth the money. Anyway, go to a sleep dot com slash lux to get $200 off and to support this podcast. This show is also sponsored by Dollar shave club. Try them out with the one time offer for only $5 and free shipping at dollar shave club dot com slash lex. The Starter Kit comes with a six blade razor refills and all kinds of other stuff that makes shaving feel great. I've been a member of Dollar shave Club for over five years and actually signed up when I first heard about them on the joe Rogan experience and now we come full circle friends. I'm interviewing joe on this podcast and monetizing it by doing a read for dollar shave club. It feels kind of surreal to be honest, to be doing a read for them, just like joe did all those years ago for the past few years. I just used the razor and the refills, but they encouraged me to try the shave butter, which I did and I love it again. Try the ultimate shave starter set today for just five bucks plus free shipping at dollar shave club dot com slash lex. And now here's my conversation with joe Rogan. Do you ponder your mortality? Are you afraid of death? [/INST] I I do think about it sometimes. I mean it does pop into my head sometimes. Just the fact that uh, I mean I'm 53, so if everything goes great, I have less than 50 years left. You know, if everything goes great, like no car accidents, no injuries, </s><s>[INST] but it could happen today. This could be your last day, that's kind of a stoic thing to meditate on death. There's a, there's a bunch of philosophers, Ernest Becker and uh Sheldon Solomon, they believe that death is the at the core of everything I wrote this book, Warm at the core. So does that come into play and the way you see the world? [/INST] I think having a sense of urgency is very beneficial and understanding that your time is limited can aid you greatly. I think knowing that this is a temporary time that we we have finite life spans. I think there's a, there's great power in that because it motivates you, it gets you going, I think being an immortal, living forever would be one of the most depressing things, particularly if everybody else was dying around you. And I think one of the things that makes life so interesting and fascinating is that it doesn't last. You know, that you you really get a brief amount of time here and really by the time you're just starting to kind of figure yourself out who you are and how not to screw things up so bad. Like time's up </s><s>[INST] the rise over what about from your, like from your daughter's perspective, do you do you think about the world we're in now and what kind of world you're going to leave them? I do. Do you worry [/INST] about it? I do, yeah, I do. I do when I see these protests and riots and chaos and so much, so much anger in the world today, and then particularly today, I think because of the pandemic and the fact that so many folks are out of work and through no fault of their own and can't make ends meet and just people feel so helpless and angry, it's a particularly divisive time, it's a particularly turmoil filled time and uh it just doesn't seem like the world of a year ago, even just feels very chaotic and dangerous and this and it's a small thing, like in terms of the, like the possibilities of things that could happen to the world, like a pandemic, like the one we've experienced, it really just doubles the amount of deaths on a bad flu year. So it's relatively speaking, is a small thing comparison to super volcano eruptions, asteroid impact, a real horrific pandemic or one that, you know, really wipes out millions and millions of people. It's um, it's stunning how fragile civility is. It's stunning how fragile our our society really is that something like this can come along, some unprecedented thing, unprecedented thing can come along and all of sudden everybody's out of work for six months and then everybody's at each other's throats and then politically everyone's at each other's throats. And, and then with the advent of social media and uh, the images that you can see, you know, with the videos of police abuse and just racial tensions are at an all time high. To a point where like if you asked me just five or six years ago, like our have racial problems in this country largely been alleviated. I'd probably say, yeah, it's way better than it's ever been before. But now you could argue that it's not now. You could argue it's no, it's way worse and in just a small amount of time, it's way worse than it's ever been during my lifetime, while while I'm aware of it, You know, obviously when I was a young boy in the sixties, they were still going through the civil rights movement, but now it just seems very fever pitched? And I think a lot of that is because of the pandemic and is because of all the heightened uh just tension. What I liken it to is um Pete road rage because you know, people have road rage, not just because they're in the car and no one can get to them, but also because you're at a heightened state because you're driving fast and you know, you're driving fast, you know, you have to make split second movements. And so anybody doing something like what people go crazy because they're they're already at an eight because they're in the car and they're moving very quickly. That's what it feels like with today with the pandemic, feels like everybody is already at an eight. So anything that comes along, it's like light it all on fire, you know, burn it down. Like that's part of what I think is part of the reason for a lot of the looting and the riots and all the chaos. It's not just the people out of work, but it's also that everyone feels so tense already and everyone feels so helpless and it's like, you know, doing something like that makes people just, yeah, it gives people a whole new motivation for chaos, a whole new motivation for, for doing destructive things that I've never experienced in my life and </s>

<s>[INST] History doesn't matter. You're exploring ancient history and drawing deep wisdom from it and at the same time saying we need to forget all of it. We need to put that behind us. [/INST] Absolutely. We need to forget it and we need to move forward, </s><s>[INST] There's so much wisdom to draw from history, even the 20th century. Look at this communism without the lessons of history of the 20th century, communism sounds like a great idea except that some of the worst atrocities conducted by Stalin and mao that killed 500 plus million people, not just killed, tortured, starvation. Where people, the cannibalism, they ate each other, they ate their Children. There's, there's just dark people. People should read some books on this on the holiday more in 1930's [/INST] with that </s><s>[INST] lesson we would and now it's becoming more popular Marxism and communism [/INST] disagreeing that that happened. I'm disagreeing that we need to harp on the things that happened because the truth is I'm giving a fact 50% today of, let's say you don't call black people do right. Black people's deaths today is abortion today right now like that. It's not racism that's too wide of a term. It's genocide and population control that boy black people are in today in America that is promoted by the music and the media that black people make that jewish record labels get paid off of our media companies, record labels, media company also. So let's give it a wider, you </s><s>[INST] know, I agree with martin Luther king. I have a dream to [/INST] that one day. You </s><s>[INST] would not be judged by the color of your skin or your race, black or jew. But by the content of your character, all the assholes that fucked you over in the music industry, Fuck artists over in the music industry are individuals. They're not jews. Can you, can you [/INST] say they are? They are jewish. </s><s>[INST] It doesn't their human with opportunities and they took those opportunities. I don't care [/INST] if they're jewish. You feel like I should release that pain and separate it. </s><s>[INST] Yes. [/INST] Okay. So that, okay, so if you're saying I should release that pain and separate it, then I'm telling you, you should release your pain and separating and we could get to this the list of how you are being with engineering opportunities. </s><s>[INST] That's 100%. I see what you're doing that exactly. You're doing the pain. I'm gonna let it go. Engineering challenges. I get it. One of the problems you highlight is people get sucked over in the, in the music industry and get sucked over in the media, get sucked over all over the place they [/INST] created. There was a, there was a jewish trainer that brought me to the hospital and put in press that I went to the hospital. I got, I know friends that off of exhaustion, a jewish doctor they diagnosed because they were right, diagnosed me with bipolar disorder and shot me with medication and put me on medication then put it in the press and every time even if I wore the wrong color hat that a *** is not supposed to wear, right? Then they immediately say he's off of his ship. He's off his meds, he's off his rocker and it's literally used as a scarlet letter control mechanism for the people understand the kids, the colleges, the high schools, What do you think they put me right now? They put me as the prophet, not the leader. It doesn't have to be the leader, right? Because we, we need, we need a more intelligent person to be the leader, but at least right. They put me as the prophet. They put me as the only person That would say this. And I'm just saying that was four Jewish members that controlled my voice because for the fact that 90% of black people in entertainment, from sports to music to acting are in some way tied into Jewish business people, meaning that in some way just like if if rom is sitting next to Obama or Jared sitting next to trump, there's a jewish person right there controlling the country, the jewish controlling that, who gets the best video or not controlling what the media says about me, </s><s>[INST] not jewish, but [/INST] they are though. That's the only thing. It just so happens that they are, it's just </s><s>[INST] saying that [/INST] doesn't mean that I hate them. That just means that they are, </s><s>[INST] but it's a, it's a dog whistle to let me, let me just say as I would love to add more love to the world. I would love you to do that as a person with a big voice with a big powerful voice that a lot of people look up to. And when you say jewish media, it's funny how this world works that way. When you say jewish media or jews are controlling the voice of black artists, black people, black artists. When you say [/INST] that you </s><s>[INST] can say it, there's a large number of people that are hurting and have anger and even have hate in their heart. When they, when they hear jewish media, they start that hate starts being directed towards the jewish people. Do you acknowledge that? That's do you understand, can you feel the hate in the world? That's when that uh that comes to the surface when you say stuff like that. [/INST] Okay, I I feel that when we go into therapy, my dad was a therapist and obviously I've got some therapy skills myself, a lot of people. My music is healing these type of spot like existences or healing the color palettes that use our healing and we can use healing words. Right? What I really feel there, I feel that there's no accountability and no responsibility on jewish people in media to at least start with owning up with the facts of what's dealt with and tell me if I say the facts out loud to the point of ari emanuel writing a letter in the financial times, trying to take food out of my Children's mouth, telling people that they're not allowed to work with me, even chris uh, como or pierce morrigan. Getting me to apologize and separate jewish business people to from the families of the jewish business people, which I did update. I did that that already happened and </s>

<s>[INST] But everything you've described now maybe you can correct me but it kind of speaks to the materialistic nature of the hardware. [/INST] But even if you add magical ectoplasm software, you didn't produce that either. </s><s>[INST] I I know, but if you can think about the actual computation running on the hardware and running on the soft, there's something you said recently which you think of um culture is an operating system. So if we just remove ourselves a little bit from the conception of human civilization being a collection of humans and rather as just being a distributed ah computation system on which there is some kind of operating system running and then the computation that's running is the actual thing that generates the interaction of the communications. And maybe even free will the experiences of all those free will. Do you ever think of? You ever try to reframe the world in that way where it's like ideas are just using us, thoughts are using individual nodes in the system and they're just jumping around and they also have ability to generate like experiences so that we can push those ideas along and basically the main organisms here are the thoughts not the humans. [/INST] Yeah but but then that erodes the boundary between self and world. Right? So so then there's no self really, really integrate itself to have any kind of will at all. Like if you're just a meme plex. I mean, if there's just like if you're just a collection of memes, and I mean, like, we're all kind of like, like currents like eddies in in this river of of ideas. Right? So, it's like, um and it seems to have structure, but there's no real boundary between that part of the flow of water and the rest. I mean, if if our and I would say that much of our mind answers to this kind of description. I mean, so much of our mind has been, it's obviously not self generated, and it's not it's not going to find it by looking in the brain. It's it is the result of culture largely. But also, um uh, you know, it's it's the genes on one side and culture on the other meeting too allow for um, manifestations of mind that don't that aren't actually bounded by the person in any clear sense. It was just the example I often use here, but there's so many others is just the fact that we're following the rules of english grammar to whatever degree we are. It's not that we certainly haven't consciously represented these rules for ourselves. We haven't invented these rules. We haven't I mean, there are there are norms of language use that we couldn't even specify because we haven't we're not grammarians we're not, we haven't studied this, we don't even have the right concepts. And yet we're following these rules and we're noticing, you know, we're noticing as you know, an error when we fail to follow these rules. Um and virtually every other cultural norm is like that. I mean, these are not things we've invented. You can you can consciously decide to to scrutinize them and override them. But I mean, just think of just think of any social situation where you're with other people and you're behaving in ways that are culturally appropriate, right? You're not being, you know, you're not being wild animals together. You're following, you have some expectation of how you shake a person's hand and how you um deal with with implements on a table, how you have a meal together. Obviously this can change from culture to culture and and people can be shocked by how different those things are. Right? We, you know, we all have foods we find disgusting. But in some countries, dog is not one of those foods, right? And and you know, you and I presumably would be horrified to be served dog. Ah those are not norms that were they are outside of us in some way and there yet there there felt very viscerally, I mean, there certainly felt in their violation. You know, if you're if you are just imagine you're in somebody's home, you're eating something that tastes great to you And you happen to be in Vietnam or wherever, you know, you you didn't realize dog was potentially on the menu and you find out that you've just eaten 10 bites of what is uh, you know, really a cocker spaniel and you feel this instantaneous urge to vomit right? Based on an idea, right? Like, so like you did not, you're not the author of that norm that gave you such a powerful experience of its violation. And I'm sure we can trace the moment in your history, you know, vaguely where it sort of got in very early on as kids. You you realize you're treating dogs as pets and not as food or as potential food. Um, but yeah, no, it's but the point you just made opens us to like we are totally permeable to a sea of mind. </s><s>[INST] Yeah. But if we take the metaphor of the distributed computing systems, each individual node is as part of performing a much larger computation. But it nevertheless is in charge of doing the scheduling of So assuming it's Lennox is doing the scheduling of processes and it's constantly alternating them. That note is making those [/INST] uh, </s><s>[INST] choices that no sure as hell believes it has free will and uh actually has free will because it's making those hard choices, but the choices ultimately are part of much larger computation that it can't control, Isn't it possible for that note to still be that human node is still making the choice [/INST] Well, it is. So I'm not saying that your body isn't doing really doing things right? And some of those things can be conventionally thought of as choices, right? So it's like I can I can choose to reach and it's like it's not being imposed on me, that would be a different experience. So there's a there's an experience of of all, you know, there's there's definitely a difference between voluntary and involuntary action. There's uh so that has to get conserved by any account of the mind that jettisons free. Will you still have to admit that there's a difference between a tremor that I can't control and a purposeful motor action that I can control and I can initiate on demand and its associated with intentions. Uh and I it's got different uh you know, motor copy which which is being which is being predictive so that I can notice errors. You know, I have expectations when I reach for this. If my hand were actually to pass through the bottle, because it's a hologram, I would be surprised, Right? And so that shows that I have a expectation of just what my grasping behavior is going to be like even before it happens. Whereas with a tremor you don't have the same kind of thing going on. That's a distinction we have to make. Um So I am, yes, I'm really the process of my intention to move subject, which is in fact can be subjectively felt really is the proximate cause of my moving, it's not coming from elsewhere in the universe. I'm not saying that. So in that sense, the note is really deciding to execute, you know, the the subroutine now, but um that's not the feeling that has given rise to this conundrum of free will, right? So that people feel like people feel like they crucial things that people feel like they could have done otherwise, right? That's the that's the thing. That's so so when, when you, when you run back the clock of your life, right? Run back the movie of your life, you flip back the few pages in the novel of your life, they feel that at this point they could behave differently than they did, right? So, like I like, but given, you know, even given your distributed computing example, it's either determine a fully deterministic system or it's a deterministic system that admits of some random, you know, influence. In either case, um that's not the free will. People think they have the free will. People think they have is damn, I shouldn't have done that. I I just like, I shouldn't have done that. I could have done otherwise, right? I should have done otherwise, right. Like, if I, like, I, like, if you think about something that you deeply regret doing right or that you do you hold someone else responsible for because they really are the upstream agent in your mind of what they did. You know, that's an awful thing that that person did and they shouldn't have done it. There is this illusion and it has to be an illusion because there's no picture of causation that would make sense of it. There's this illusion that if you arrange the universe exactly the way it was a moment ago, it could have played out differently. And the only way it could have played out differently is if there's randomness added to that. But randomness isn't what people feel would give them free will if you tell me that, you know, I I only reach for the water bottle this time because somebody is because there's a random number generator in there, kicking off values and it finally move my hand. That's not the feeling of authorship </s>

<s>[INST] So you mentioned a couple of times, uh, this fascinating world of entrepreneurship in china of the fearless nature of the entrepreneur. So can you maybe talk a little bit about what it takes to be an entrepreneur in china? What are the strategies that are undertaken, What are the ways to achieve success? What is the dynamic of the CIA funding of the way the government helps companies and so on. What are the interesting aspects here that are distinct from, they're different from the Silicon Valley world of entrepreneurship. [/INST] Well, many of the listeners probably still with brand chinese entrepreneur as copycats And no doubt 10 years ago, that would not be an inaccurate description Back 10 years ago, an entrepreneur probably could not get funding if he or she could not describe what product he or she is copying from the U. S. Um The first question is who has proven this business model? Which is a nice way of asking for copying and and that reason is understandable because china had a much lower internet penetration and and I didn't have enough um indigenous experience to build innovative products. And secondly internet was emerging. Link startup was the way to do things, building the first minimum viable products and then expanding was the right way to go. And the american successes have given a shortcut that if you took your, if you build your minimally viable product based on an american product, it's guaranteed to be a decent starting point, then you tweak it afterwards. So as long as there's no I. P. Infringement, which as far as I know there hasn't been in the in the mobile and Ai spaces, um that's a much better shortcut and I think Silicon Valley would view that as still not very honorable because that's not your own idea to start with, but you can't really at the same time, believe every idea must be your own and believe in the lean startup methodology because lean startup is intended to try many many things and then converge one that works and it's meant to be iterated changed. So finding a decent starting point without legal violations, there should be nothing morally dishonorable about that. </s><s>[INST] So just a quick pause in a, it's fascinating that that's uh, why is that not honorable? Right is exactly as you formulated is it seems like a perfect start for business is to, uh, to take, you know, look at amazon and say, okay, we'll do exactly what amazon is doing. Let's start there in this particular market and then lets out, innovate them from that starting point come up with new ways. I mean, is it wrong to be? Uh, except the word copycat. It sounds bad, but is it wrong to be a copycat? It just seems like a smart strategy. But yes, doesn't have a heroic nature to it. That, uh, like a, like a steve jobs, um, Elon musk sort of in something completely coming up with something completely new. Yeah, [/INST] I like the way you describe it is a non heroic, uh, acceptable way to start the company and maybe more expedient. So that's the, that's I think a baggage for Silicon Valley that if it doesn't let go, then it may limit the ultimate ceiling of the company take Snapchat as an example, I think, you know, Evans brilliant, he built a great product, but he's very proud that he wants to build his own features, not copy others while facebook was more willing to copy his features and you see what happens in the competition. So I think putting that handcuff on the company would limit its ability to reach the maximum potential. So back to the chinese environment, copying was merely a way to learn from the american masters. Just like we if you would, we learn to play piano or painting, you start by copying, you don't start by innovating when you don't have the basic skill set. So very amazingly the chinese entrepreneurs about uh six years ago started to branch off with these lean startups built on american ideas to build better products than american products. But they did start from the american idea and today we we chat is better than WhatsApp web, oh is better than twitter who is better than cora and so on. So that I think is um um chinese entrepreneurs going to step two and step three is once these entrepreneurs have done one or two of these companies, they now look at the chinese market and the opportunities and come up with ideas that didn't exist elsewhere. So products like and financial under which includes alipay which is mobile payments and also the financial products for loans built on that and also in education V. I. P. Kid and in uh social social video, social network Tiktok and in social e commerce pin dodo and then in ride sharing mobile like these are all chinese um innovative products that now are being copied elsewhere. So and then an additional interesting observation is some of these products are built on unique chinese demographics which may not work in the US but may work very well in Southeast Asia Africa and other developing worlds that are a few years behind china and a few of these products may be are universal and are getting traction even in the United States such as Tiktok. So this whole ecosystem is supported by VCS uh as a virtuous cycle because a large market with with the innovative entrepreneurs will draw a lot of money and then invest in these companies as the market gets larger and larger U. S. Market china market is easily three or four times larger than the U. S. Um They will create greater value and greater returns for the VCS, thereby raising even more money. Um So That's innovation ventures, our first fund was 15 million. Our last fund was 500 million. So it reflects the valuation of the company's and R US going multi stage and things like that. It also has government support uh but not in the way most americans would think of it. The government actually leaves the entrepreneurial space as a private enterprise sort of self regulating and the government would build infrastructures that would um around it to make it work better. For example, the mass entrepreneur, mass innovation plan Build 8000 incubators. So the pipeline is very strong to the VCS for autonomous vehicles. The chinese government is building, uh, smart highways with sensors, smart cities that separate pedestrians from cars that may allow initially an inferior autonomous vehicle company to launch a car without increasing with lower casualty because the roads are, the city is smart and the chinese government at local levels would have these guiding funds acting as LPS, passive LPS to funds. And when the fund makes money, part of the money made is given back to the GPS and potentially other LPS to increase everybody's return at the expense of the government return. So, that's interesting incentive that entrusts the task of choosing entrepreneurs to VCS who are better at it than the government by letting some of the profits? I'll move that way. </s>

<s>[INST] So, but nevertheless, if we looked at our solar system, it would look and I could see every single object would look like a sphere. [/INST] The inner part where the planets are would look like, you know, flat, right? The Kiper belt and the asteroid belt have a larger um, it </s><s>[INST] gets fatter and fatter. [/INST] It's kind of like atmosphere. That's right. And if you look at the very outside it's polluted by this, you know, quasi is for royal thing. Nobody is of course ever seen the ord cloud. Right? We've only seen comments that come from the old cloud. So the word cloud, which is this right population of distant debris its existence. It's also inferred, you could say, alternatively, there is, you know, there's a big cosmic creature that occasionally, You know, sitting at 20,000 au and occasionally throws an icy rock towards the sound like a </s><s>[INST] monster I think it's called. Okay, I mean, so it's a mystery in many ways, but you can kind of in for a bunch of things about it, by the way, both terrifying and exciting that there's this vast darkness all around us that's full of objects that they're just throwing just there. [/INST] Yeah, it's actually kind of astonishing, right, that we have only explored a small fraction of the solar system. Right. That that really kind of baffles me because I remember as a student, you know, studying physics, you know, you do the problem where you put the earth around the sun, you solve them like it's one line of math and you say, okay, well that surely was figured out by Newton. So like all the interesting stuff is not in the solar system, but that um, it's just plainly not true. There are, there are mysteries in the solar system that are remarkable that we are only now starting to just kind of scratch the surface of </s><s>[INST] and some of those objects probably have some information about the history of our solar system. [/INST] Absolutely. Absolutely. Like a great example is uh, you know, small media writes, writes, small meteorites are melted, right? They have their differentiated meaning some of the uh, iron sinks the corn you say, How can that be? Because they're so small that they wouldn't have melted just from the heat of their accretion. Turns out the fact that the solar nebula, the disk that made the planets was polluted by aluminum 26 is in itself a remarkable thing. It means the solar system did not form in isolation. It formed in a giant cloud of thousands of other stars that were also forming some of which were undergoing, going through supernova explosions, some of and releasing these unstable isotopes. Um That of which we now see kind of the traces of it's so cool. </s><s>[INST] Do you think it's possible that life from other solar systems was injected and that that was what was the origin of life on earth? [/INST] The panspermia idea, </s><s>[INST] um, that's seen as a low probability event by people who studied the origin of life. But that's because uh then they will be out of a job. [/INST] Well, I don't think they would be out of the job because you just then say, you have to figure out how life started there. </s><s>[INST] But then you have to go there. We can study life on Earth much easier. We can study in the lab much easier because we could replicate conditions that are from an early Earth. Much easier from a chemistry perspective from a biology perspective, you can intuit a bunch of stuff, you can look at different parts of earth. And just [/INST] to an extent, I mean, the early Earth was completely unlike the current earth, right? There was no oxygen. So, uh, one of my colleagues at caltech joker schnick is, I'm certain my, Something like 100% certainty that life started on Mars and came to earth in on Martian meteorites. Uh, this is not a problem that I like to kind of think about too much like the origin of life. It's a fascinating problem, but you know, it's not physics and I just like, just don't, don't love it. </s><s>[INST] It's the same reason you don't love, I thought you're a musician. So music is not physics either. So [/INST] Why, Why are you selling? 200% physics? Okay. No. Look, in all seriousness though, I, uh, there are a few things that I really, really enjoy. I genuinely enjoy physics. I genuinely enjoy music. I genuinely, you know, enjoy martial arts and I genuinely enjoy uh, my family, I should have said that all in a reverse order or something, but I like to focus on these things and not worry too much about about everything else. You know what I mean? Just because there is a, like you said earlier, there's a time constraint you can't do it all. </s><s>[INST] There's many mysteries all around us. So, uh, and they're all beautiful in in different ways to me, that thing I love is artificial intelligence that perhaps I love it because eventually I'm trying to suck up to our future overlords. The the question of you said there's a lot of kind of Little pieces of evidence for this thing. This planet nine. If we were to try to collect more evidence or be certain like a paper that says like you drop it clear or done, What does that require? Does that require us sending probes out or do you think we can do it from telescopes here on Earth? What are the different ideas for conclusive evidence for planet 9? [/INST] The moment planet nine gets imaged from a telescope on Earth. It's done. I mean, it's just there. </s><s>[INST] Can you clarify because you mentioned that before from an image, would you be able to tell? [/INST] Yes. So from an image, the moment you see something, something that is reflecting sunlight back at you and you know that it's hundreds of times as far away from the sun as as as the Earth you're done. So </s>

<s>[INST] more like a legal type of uh, get the details of where he was, what he was thinking. What I [/INST] think it's like, Yeah, I think it's, I think the public probably cares to get to know what happened to and, and again, I think if you're, if you're careful, you can expose someone for as they lie to you without giving in to those lies, right? Like without capitulating to, oh, I'm just gonna assume you're correct. I think you can point to, well, lex you say it happened this way, but you've lied about X, y and z. Why should we believe you? That's a suddenly a totally different conversation than just being like, okay, that's how it happened. </s><s>[INST] The thing I caught that bothered me and the thing I hope to do in interviews of um, if I eventually get good at this thing is the human aspect of it, which is like, which I think you have to do in person is he seems a bit nonchalant about the pain and the suffering of people that I have red flags about in the way he's, he communicates about the loss of money, like the pain that people are feeling about the money. I get red flags. Like you're not uh, forget if you're involved in that pain or not, you're not feeling that pain. [/INST] Well he'll he'll say he is but he'll be playing a game of League of Legends while doing it. </s><s>[INST] No, but I just see from his face [/INST] that I know and that needs to </s><s>[INST] be grilled like that little human dance there. That's what I mean. Um I considered I talked to him, I considered doing an in person interview with him. Um And [/INST] are you still considering it? I </s><s>[INST] don't know. Do you think do you think I should in person? I [/INST] think it depends if you think you have anything to add to the conversation. A lot of people have </s><s>[INST] been already. You did an incredible job. [/INST] Thanks. I think </s><s>[INST] I think I would like to grill the shit out of him as a fellow human but not investigative Coffeyville investigator, like like another human being, another human being who I can have compassion for who has caused a lot of suffering in the world like that that grilling like basically convey the anger that people and the pain that people are feeling right like that. [/INST] I think I think it'd be really hard. I mean like that guy is sort of a master dancer and what he would say at the end of it because I've listened to so many interviews of him. Um I probably like a gpt model for Sam I think he would do some kind of thing about like Yeah, I really um I really hear you and um you know it's it's it's just terrible. I feel such an obligation to the people who lost money and you know, it's just, it's a lot of money, it's a lot of money, you know, he'd do something like that and it would be very superficially like, okay, but when you, when you drill down to the details of what he did, it's just impossible that he didn't know. And one of the things that I wish I had asked, maybe I can talk about, like, I wish I had gone on this. It's just so hard when you're doing a live interview to kind of focus on one thing everyone's asked about the terms of service. So in the terms of service, there was like, we can't touch your funds, your funds are safe, we're never gonna do anything with that anytime anyone brings that up, he says, oh, well there's this other terms of service over here with margin trading accounts, remember we talked about, it's a derivatives platform, if you're in our derivative side, you have your subject to different terms of service, which kind of lets us like move your money around with everyone else. Okay, so we treat it as one big pool of funds and that's sort of, the explanation of how this all happened is we had this huge leverage position and we lost everything, but what no one has sort of done a good enough job getting to the heart of, is that this pool of funds never was segregated properly. It was all treated under the same umbrella of we can use your funds, There was no amount of we have the client deposits which were just deposited with us and not like used to margin trade or do anything over here. These funds over here, we have saved, they didn't fundamentally they lied from the get go about how they were treating the most precious assets, which is your customer deposits that you said you didn't invest. Clearly you put them all over here, you YOLO gambled them. And then when everyone starts withdrawing from here, they don't have any money over here. So that is like one of the most fundamental things that I haven't seen anyone grill him on and uh the next time if I get the chance to ambush him again, that's that's what I'm going to drill </s><s>[INST] down on because it's [/INST] impossible for that not to be fraud. There's no world where you had a pool of funds over here and now you don't have them without you somehow borrowing over here because if you deposited one Bitcoin and I never sold that Bitcoin and its earmarked lex Friedman and you come and it's not there something had to happen. Right. </s>

<s>[INST] conscious self aware exactly to us it does make me a little bit sad as a human and just watching all the breakthroughs on artificial intelligence side when applied to natural sciences now more more to physics that the creatures that will solve the question of the origin of the nature of life or just the process the nature of life will be A. I. systems makes me a bit sad too [/INST] I don't think so </s><s>[INST] because you think that humans will [/INST] at this point in time remember was behind a I you know I I'm not buying in the singularity thing yet two A. I. is not aware A. I. is being built by humans so A. I. is a tool and extremely smart tool as long as we build it and as long as we use it as a tool it remains a tool and I think there is a lot of brew ha ha and of course the science fiction and movies they don't help </s><s>[INST] check I got to push back a little bit yes I agree with you for the most part in terms of ha ha ends scifi but there is I can work of DeepMind we can look at Chesil economic approaching folding so chess is a simple one to first look at what are Alfa zero which is their game playing engine was able to discover Nostoc fish about chess humbles the best human players not just is better than them it comes up with ideas that the humans don't understand and so the A. I. knows telling you even those programmed by humans the allies saying like sacrificing a upon here is a good idea sacrificing a queen bishop here's a good idea and then you start to kind of into it as a human why we don't deeply understand and you can say that a I was not conscious it doesn't deeply understand what humans do but there still are a wisdom and a depth of knowledge in that chess playing program that humans don't have in the same with the alfalfa folded protein folding there's a another planet to physics to sis simulating nuclear reactions and so on it feels like there might be a way to understand the nature of life that we can kind of into it poetically as humans but the the true understanding will come from a system that's much welcome competition sophisticated [/INST] again you know I would push back on on my turn because I still think that human give themselves the ability to do that by building that tool so the idea that the tool you know </s><s>[INST] what what what [/INST] we are getting into the car she scale and and and and the dark forest and all these things we can see the world this way I at this point in time for me I still see the great tool now are whether this I. F. I see are you always going to you know happened it's a try I still think that we are far away from from this but if that tool is capable of giving me a new perspective it's just that we're starting to jump into a deeper condition of of what the universe is whether it's through our brain though through a different way of gathering information remember this is what we do </s><s>[INST] yeah humans able to actually build tools and then like integrate them enter their way of thinking maybe another generation has to be born that is raised with those tools but we seem to like take for granted all the cool technologies you're integrate into your way of thinking a lot of people are growing up now their mind is integrated with the internet and yet you basically reconfigure the way you memorize things you no longer have to memorize a lot of fast casual look come up really quickly [/INST] yeah </s><s>[INST] and so like S. U. re allocate a lot of resources for thinking versus memory of just strict facts that kind of stuff we integrate all of that [/INST] yeah and and you know they are I would completely agree with you in fact I wrote about this again in in this new book that's coming out </s><s>[INST] what is the book coming up [/INST] in January it's evil been French actually just off with </s><s>[INST] you wrote in French [/INST] actually a rooted first in English </s><s>[INST] yes [/INST] and I translated into French so the the English version is already pretty much ready to go if we find a publisher in yet in the US but anyways the point of being here is that I looked at this as our relationship with technology as a complete change to me this is the singularity more than anything else which she is the co evolution of human with technology not anymore with their environment why we are missing of the environment right now why we don't respond to when they make the way we should because we are disconnected to the environment we are taking our information from and we were adapting from right now exactly as you said we take the information from the web from the phones it's a truck we have no filter over that information before you were out in the environment the information you get is the one the planets is sending you now this information is coming from different way you have no way of knowing if for missions correct or not you're gonna push back on that now yeah you look as this is an echo system and it explains a lot about behavior </s>

<s>[INST] So that that significantly reduces the mental effort or significant increases the efficiency of the teacher. The human [/INST] teacher. Exactly. And it makes more or less anyone able to to be a teacher in that um in that way, so that's that's part of the the answer. And then the other is that the system on its own will be able to um through reading through conversations with other people and so on. Um learned the same way that you or I or some other humans do. </s><s>[INST] First of all, that's that's a beautiful vision. Um I'll have to ask you about semantic weapon in the second year, but first uh [/INST] are </s><s>[INST] there when we talk about specific techniques, do you find something inspiring or directly useful from the whole space of machine learning? Deep learning, these kinds of spaces of techniques that have been shown effective for certain kinds of problems in the recent? Uh no, decade and a half [/INST] I think of machine learning work um as more or less what our right brain hemispheres do. So um being able to um take a bunch of data and recognize patterns being able to statistically in for things and so on. Um and um you know, I certainly wouldn't want to not have a right brain hemisphere, but I'm also glad that I have a left brain hemisphere as well, something that can metaphorically sit back and puff on its pipe and think about um this thing over here, it's like why might this have been true? Um and um what are the implications of it? How should I feel about that and why and so on? So um thinking more deeply and slowly. Um um what can a man called thinking slowly versus thinking quickly? Whereas you want machine learning to think quickly, but you want the ability to think deeply even if it's a little slower. Um so I'll give you an example of a project we did recently with NIH involving the Cleveland clinic and um a couple other um institutions that we ran a project for. Um and what it did was it took um glasses, genome wide association studies. Um uh Those are sort of big databases of patients that came into a hospital. They got their D. N. A sequence because the cost of doing that has gone from infinity to billions of dollars $200 or so. Um And so now patients routinely get their D. N. A. Sequence. So you have these big databases of the snips, the single nucleotide polymorphisms, the point mutations in a patient's DNA. And the disease that happened to bring them into the hospital. So now you can do correlation studies, machine learning studies of which mutations um are associated with and lead to which physiological problems and diseases and so on. Like getting arthritis and um and so on. And the problem is that those correlations turned out to be very spurious. They turned out to be very noisy. Um Very many of them have led doctors on to wild goose chases and so on. And so they wanted a way of eliminating or the bad ones are focusing on the good ones. And so this is where psych comes in which is psych takes those sort of a dizzy correlations between point mutations and um the medical condition that needs treatment. Um And we say okay let's use all this public knowledge and common sense knowledge Um about what reactions occur where in the human body um what prelim arises what what catalyze is what reactions and so on. And let's try to put together a 10 or 20 or 30 step causal explanation of why that mutation might have caused that medical condition. And so sick would put together in some sense some rube Goldberg like um chain that would say oh yeah that um mutation um if it got expressed would be this um um altered protein. Which because of that if it got to this part of the body would catalyze this reaction. And by the way that would cause more bio active vitamin D. In the person's blood. And anyway 10 steps later that screws screws up bone resort option. And that's why this person got osteoporosis early in life and so on. </s><s>[INST] Human interpret herbal or at least dr human. [/INST] Exactly. And um the important thing even more than that is um uh you shouldn't really trust that 20 step um uh rube Goldberg chain any more than you trust that initial a dizzy correlation except two things. One if you can't even think of one causal chain to explain this. Um Then that correlation probably was just noise to begin with. And secondly. And even more powerfully along the way that causal chain will make predictions like the one about having more bio active vitamin D. In your blood. So you can now go back to the data about these patients and say by the way did they have slightly elevated levels of bio active vitamin D. In their blood and so on. And if the answer is no, that strongly dis confirms your whole causal chain, if the answer is yes, that somewhat confirms that causal chain. And so using that we were able to take this these correlations from this glass database and we were able to um essentially focus the doctors, focus the researchers attention on the very small percentage of correlations that had um some explanation and even better some explanation that also made some independent prediction that they could confirm or discomfort um by looking at the data. So think of it like this kind of synergy where you want the right brain machine learning to quickly come up with possible answers. You want the left brain psych like ai to um you know, think about that and I like think about why that might have been the case and what else would be the case if that were true and so on and then suggest things back to the right brain to quickly check out again um to um So it's that kind of synergy back and forth. Which I think is really what's going to lead to general ai not um narrow brittle machine learning systems and not just something like psych </s>

<s>[INST] You have to. That comes from being able to listen to some inner voice. [/INST] It's not gonna come from elsewhere. It's not gonna come from elsewhere. Yeah, I guess. It's really hard, because you're living the life and there's things happening around you and people tell you what to do and what not to do and all. No one really has their own opinions. Everyone is kind of like listening to the cooler thing, it's never really- Except Steven, he seems to stand on his own. Okay, so Charles. So we think. Yeah, I'd say so, and like... High Hashtag. Something I realized too, like, cause we just went to TwitchCon and we were talking to a lot of streamers. What was that? It was interesting. I thought it was interesting because the few people that I feel like I, that seemed really cool and that I look up to, like in the streaming world, all of them wants to quit streaming. All them wants to do it. No one wants like, no one likes that. And they're so successful. Like they are around successful people. They're working every single day, they're working hard, they're making so much money. And everyone is just complaining and like they're complaining about like not being able to see their partner or you know, because they need to live somewhere else. Because I see these things and they seem extremely unhappy. And but it's so hard for them to just like cut all this successful stuff off because that's like what you, you know, learned to do. And that's like supposed to be like your happiness, but it isn't. Everyone is really unhappy. </s><s>[INST] Except Steven, he seems to stand on his own. Totally. Yeah, I'd say so. I like the hashtag. How was that? Yeah, there's something about maybe streaming is different, but YouTube folks too have interacted with a few and even in podcasting space, people become obsessed about the views and numbers and subscribers and stuff like that. So I turn, I never talk about that. I don't pay attention to that. I feel like that's a drug that destroys your mind. Your mind is an artist, the ability to create truly unique things. Also your mind in terms of the anxiety, the ups and downs of the attention mechanism. And then also being just, if something that you make is not popular, but it meant a lot to you, you will think of it less because it's not popular. That's a really dangerous thing. And because everyone around you is reinforcing, like I'll get messages like, wow, this thing got this many views or something. Great job. It's like, no, you don't get it. That's not, everyone is enforcing this language of views and likes and so on. And it's correlated of course, because truly impactful things will get a lot of attention often, but it's not, on the individual local scale, like temporarily, it can really fuck with your mind. I see that in the creators, they become addicts. [/INST] For sure. And to the algorithm. Lost in chasing views. We know friends that we know cool people and then they start streaming and eventually they're chasing the dragon of like. Yeah. And they change, yeah. It's like hard to engage for them. To change who they are clearly. Hard to engage for them anymore. And there's like, this is something I've always said that one of the biggest blessings and biggest curses of humanity is we are very good at acclimating. You can become paralyzed, you can have all sorts of horrible things happen to you and get used to it, and you'll be okay. You're like a good bass-line. But it works the other way too in that you can get more and more and more and more and you acclimate to it almost immediately. There's like, this is a phenomenon that, I bet it happens in the YouTube world, but I know it happens in the streaming world where you're streaming. 1,000 viewers every day, huge event happens, and you blow up. And you got like 15,000 viewers for dare or two, and then it starts to go down and down and down and down and down, and then after all the drama started off, you're at like, 3,000 concurrent viewers. Now in the macro, you went You went from 1,000 to 3,000. That feels awesome. But you actually feel like shit the whole time because you're remembering when you had 10 or 15,000 and now everything feels horrible. And you'll see people climb over time. They're like, fuck, but whatever that one huge stream I had, I've never been able to... And it's like, dude, you're doing great. Yeah, that happens a lot. There's so many people that we know that we find super, super cool. They're passionate about things. They have so much interest and then they just get so addicted to these numbers and everything is just ruined. All the cool things about them is ruined because they stop doing the things that they actually like to do something else that gives them more viewers and more money. And it's really sad to see. </s><s>[INST] Yeah, that temporary sacrifice that seems temporary is that it actually destroys you. Like for one time, making a choice. Because I come across those choices often, like I can do this. You can kind of know what's going to be popular and not... And you have to ask yourself that question, [/INST] like, is this going to sacrifice? Because the people are sacrificing intimate relationships, they're sacrificing time with their family, they're sacrificing time with the things that they feel good about and that they like. And that's something I kind of realized last year because I was working so much and I was just grinding and grinding and grinding. Because it was kind of new for me and then New Year's came by. And I was like, wait, what did I even do? Like the entire year. Like, I traveled to a bunch of places, but nothing actually really meant anything to me because I felt like I was just working the entire time. I felt like I was just numb through the entire year. And it was really scary. Um, like I rented, um, like a super pretty house, like for a week with my dad and my sister, cause I wanted to spend time with them, but the entire time I was just streaming and I actually didn't ever like calm down and just like chill with them. And I like, that's like time. I'll never get back. I don't, I don't give a shit about the money that I made that week, but I lost the time and like that is really important to me and I, yeah. And a lot of people are doing that. Um, and I feel like you, as you said, like, you can definitely see that in like, artists for sure. I feel like if you, if you look at like artists, like back in the sixties or seventies, I feel like things were just so much better back then. And it feels like they were actually making music that meant something to them. They were actually making art. And I feel like today everything is just kind of like, whatever is cool, whatever sells whatever, you know, sounds in a certain way, everything is kind of the same thing and everything that is very artistic and very cool is actually not that popular at all. </s>

<s>[INST] it's time to meet all the theoretical physicists. Uh so he's a string theorist. But uh his idea is that uh aliens that are sophisticated enough to travel into styles like that those kinds of distances will figure out actually ways to hack the fabric of the universe enough to have fun in other ways. Like this universe is too boring. Like you would figure out ways to create other universes or like you go outside the physics as we know it. So the reason we don't see aliens visiting us all over the place is they're having fun elsewhere. This is like way too boring. We humans think this is fun, but it's actually mostly empty space. That's no fun is happening. Like there's no fun in visiting earth for super advanced civilization. So he thinks like if alien civilizations are out there, they found outside of our current standard models of physics ways of having fun that don't involve us [/INST] and that that's probably true. But even the notion of visiting that's so literally pedestrian, you know, of course we want to go there because going there is the only thing we know we see a thing we want, we want to go there and get it. But that is probably something they've no longer got need for. I specifically don't particularly want to go to space. Sounds awful. Uh you know, none of the things I like are going to be there and I my whole work is my whole career is finding life and understanding the universe. So I care a lot but I care about knowing about it and I feel no need to go there to learn about it. And I think as we develop better tools, hopefully people will feel less and less and need to go everywhere that we know about and I would expect any alien civilization worth assault. Have developed observation tools and and tools that allow them to understand the universe around them and beyond without having to go there. This this going is so wasteful. </s><s>[INST] Yeah. So more focused on the knowledge and learning versus the colonization like the conquering and all those kinds of things [/INST] that's you know beneath them? </s><s>[INST] That's beneath the that I mean that's why do you think there's um any a hopeful search for life through philosophy and other gasses, do you um, do you think there's other alien civilizations out there first, do you think there's other life out there first, Do you think there's life in the solar system? Second, do you think there's life in the galaxy? And uh, third, do you think there's intelligent life in the solar system or galaxy outside of Earth? [/INST] So intelligent life? I have no idea. It seems deeply unlikely possible, but I'm not even sure if it's plausible. </s><s>[INST] So that's a special thing to you about Earth is somehow intelligent life [/INST] campus and it's only very briefly, probably extremely briefly. </s><s>[INST] Uh you mean like it's always going to like we're gonna destroy ourselves. Exactly, boy, [/INST] and life will continue on Earth happily. Probably more happily. Um Trees </s><s>[INST] and the dolphins will be here, I'm telling you [/INST] and the cockroaches and the incredible fun guy. You know, they'll be fine. So life on earth will be fine. It was fine before us and we'll be fine after us. So I'm not that worried about intelligent life, but I think it is unlikely even on Earth is unlikely. Out of what is it? Five billion species across the history of the earth. There's been one an intelligent one and for a blink of an eye, possibly not much longer than that. Uh, so I wouldn't bet on that at all, though, I would love it. Of course, you know, I I wanted to find aliens since I was a little girl. And so of course I initially wanted to find ones that I could be friends with and I've had to let go of that dream because it's so deeply implausible. </s><s>[INST] But see the nice and sorry to interrupt. But the nice thing about intelligent alien civilizations, they may have more biosignatures than non intelligent ones, so they might be easier to detect that that would be the hope [/INST] on earth. That's not the case, but it could be the case elsewhere. It's not the case on earth. Most of the biosignatures we have on earth are created by quite simple life. If you don't count pollution, pollution is all all the baby. So </s><s>[INST] you don't see uh polluting gasses as a as a possible. Like [/INST] I look for polluting gasses, I would love to find polluting gasses. Well, you know, I'll be worried for them. Of course, the same way I I think about my alien colleagues all the time looking at us and I'm sure they worry about our pollutions, but it would be a really good, robust, unambiguous sign of life if we found complex pollutants. So I look for those two, I just don't have any hope of finding them. I think intelligent life in the galaxy at the same time that we're looking is deeply implausible, but life I think is inevitable and if it is inevitable it is common. So I think there will be life everywhere in the galaxy now how common their life is, I think will depend a lot on whether there's life in the solar system beyond earth. So I'll adjust my expectations very much based on there being life in the solar system if there's life in the Venusian clouds. If there's life in the if they're by singers coming out of the plumes of Enceladus, if there's life on titan. </s><s>[INST] Yeah, that's right. And so yeah, plumes of Enceladus, that's the that's the Saturn one, [/INST] it's the moon that has the geysers that come out and so you can't see the under the subterranean oceans, but </s>

<s>[INST] can I ask you, uh, a pothead question, which is, uh, what is money man? So if we just kind of zoom out and look at, because let's talk to you about Cryptocurrency, which perhaps could be the future of money in general, How do you think about money? He said new MRI the vision, the goal is to to run to manage the world's money, What is money in your view? [/INST] I don't have a good answer to that, but it's definitely in my personal life, it's become more and more warped. Um and you start to care about the real thing, like what's really going on here. Um Ellen has talked about things like this, like what what is the company really? It's like it's a bunch of people who like kind of show up to work together and like they solve a problem and they might not be a stock out there that that that's trading that represents that what they're doing, but it's it's not the real thing. Um and being involved in crypto um like I was early, I put crowd sale of ethereum and uh and all these other things in different crypto hedge funds and things that I've invested in, and it's like, it's just it's just kind of like, it feels like how I used to think about money stuff, it's just it's just like totally warped um because you yeah, you kind of you stop caring about like the price and you care about like the product, </s><s>[INST] so by the product, you mean like the different mechanisms that money is exchanged, I mean money is ultimately kind of a little uh you know, one is a store of wealth, but it's, it's also a mechanism of exchanging wealth, but that like, what what wealth means becomes a totally different thing, especially with Cryptocurrency to where it's almost like these little contracts, these little agreements, these transactions between human beings that represent something that's bigger and just like cash being exchanged at 7 11, it feels like, [/INST] yeah, maybe I answer, what is finance like you, it's what are you doing when you can, when you have the ability to take out a loan, you can bring uh a whole new future into being with finance uh if you couldn't get a student loan to get a college degree, you couldn't get a college degree, like if you didn't have the money, but now, like, weirdly, you can get it with and like yeah, all you have is this like loan, which is like, so now you can bring, you can bring a different future into the world and that's how I, when I was saying earlier about if you rerun american history, economic history without these, these these things like, you know, a lot of take out loans, you're not allowed to have have derivatives, you're allowed to have money. Um it was just, it just doesn't really work and it's a really magic thing, how, how, how much you can do with finance, but kind of bringing the future forward </s><s>[INST] finances empowering. It's uh we sometimes forget this, but yeah it enables innovation, enables big risk takers and bold builders that ultimately make this world better. You said you were early in on Cryptocurrency, can you give your high level overview of just your thoughts about the past, present and future of Cryptocurrency? [/INST] Um Yeah, so my friends told me about Bitcoin and I I was interested in equities a lot and I was like well it has no NPV, it has no future cash flows. Bitcoin pays no dividends. Um So I really couldn't get my head around it like that, this could be valuable. Um and then I but I did so I I didn't feel like I was early in Cryptocurrency in fact because I was like, it was like 2014, it felt like a long time after Bitcoin. Um And then but then I I really liked some of the things that uh theory and was doing, it seemed like a super visionary thing, like I was reading something that was um that was just going to change the world when I was reading the white paper um and I like the different constructs you could have inside of a theory and that you couldn't have on on Bitcoin </s><s>[INST] like smart contracts and [/INST] Exactly yeah and even the there were, yeah even spoke about different uh different </s><s>[INST] constructions, you could have, yeah, that's a cool dance between Bitcoin and ethereum of, it's in the space of ideas, feel so young like I wonder what cryptocurrencies will look like in the future. Like if Bitcoin or ethereum two point or some version will stick around or any of those like who's gonna win out or if there's even a concept of winning out at all, Is there a Cryptocurrency that you're especially find interesting that technically financially philosophically you think this is something you're keeping your eye on? [/INST] Well I don't really, I'm not looking to like invest in cryptocurrencies anymore. Um But I I they are, I mean the there and many are almost identical, I mean there's not there wasn't too much different um between even ethereum and Bitcoin in some ways right? But there are some that I like the privacy ones and I was like I like Z cash for its like coolness. It's actually um it's it's like a different kind of invention compared to </s><s>[INST] some of the other things. Can you speak to just briefly to privacy? What is there some mechanism preserving some privacy of the universe. So I guess everything is public. Yeah, that's the problem. The [/INST] yeah, none of the transactions are private. Um And so you know even like I have some of my, I have some numerous air and you can just see it. In fact you can go to website and says like you know like ether scan and it'll say like new Marie founder uh and I'm like how the hell are you guys know this? </s>

<s>[INST] disproportionate to his work. But even on your show, you can tell even outside of the very stuff that he's not ultra careful about. He's not nuance. Yeah, he's not afraid to say something just like I would say, let me just criticize my face does not use me carelessly say something controversial, right? Like I'm not saying he doesn't go, you know, that makes him it's a very different thing than, uh, uh, somebody who on purposes says something controversial stuff. Uh, like Milo Yiannopoulos, sorry, I forgot my milo, whatever his name is. Yeah. Which is really nice to see that he's a genuine person who's thoughtful, He doesn't mean to, but he just cares carelessly seems to say things that, uh, I feel like damaged the rest of his body of work. [/INST] I can't really speak for him. But I would guess his point is once you're swimming in this kind of worldview, you're going to be anathema already. So there's no pleasing these people. So why bother trying? </s><s>[INST] Yeah, I think that's a deeply that's a that's a black pill way of seeing the world. It's not black killed at all because it's a cynical way like these people. So like it's saying that you're it's a very kind of way of thinking like I'll say whatever I want whoever comes along with me. [/INST] No, you just earlier said yourself that race, racism has been weaponized as a way to shut down conversation. So I think his perspective would be I am so outside the mainstream in my world view that I know I'm going to be called racism racist. So there's no point in trying to be nuanced because I'm already going to get the scarlet </s><s>[INST] letter. Yeah, I just disagree with that because for example, I'm one, I am one person that he turned off by his carelessness and I think I should be a good target, I should be somebody that's fair. And I'm just like he it's very convenient to think that there's ridiculous people out there, which they are who call everybody racist and sexist currently and then you can't please them. So I'm not even gonna try. No, but there's like this gray area of people that I don't listen to. The outrage, culture, whatever. I don't, This Wikipedia article means nothing to me. Like, I'm not going to right what I'm more, I'm just seeing this careless person and if he's going to be careless about uh like race like this, I feel like if I walk along with him long enough, I'm going to catch the carelessness, I'm going to lose. [/INST] Like, I'll defend your perspective better than you can. Yeah, this this </s><s>[INST] is good. I'm taking notes. [/INST] I talked to eric Weinstein after you guys talked about me on your show and I was weinstein, we had a good conversation. He invited me on his </s><s>[INST] show that would be an amazing conversation [/INST] and we got on the phone and his concern fairly, he goes, I don't want you to come on my show for the purposes of clowning me and I would never do </s><s>[INST] that. [/INST] It would never be </s><s>[INST] might not be aware of. We uh [/INST] that's why he wanted to feel me out. He's like, you know when he hears troll it can mean a lot of different things and I, we had a very conversational very much it was very clear that's not where the conversation would go. But I think when you are going to be on someone's show there is a responsibility that they're not going to have to pay a cost for having you as their guests. So if you're, if you're put off by how he wasn't that live streams are too. I did like I understand where you're coming from. I think he's very, very bright but you have a very you have a different audience than I do when you're going for something different than than I am. No, </s><s>[INST] no, like in my in just a sense of you wouldn't feel safe with him. Yeah, I wouldn't feel safe with him, but he's born alive for me. I think I would like to actually talk to him one day Alex jones has crossed the other line for me what you [/INST] can do what you can do with me tape the episode that never release it </s><s>[INST] big. No, it's one of those things will be uh when there's finally they don't make a history channel documentary about you and I and how it all went wrong, like the cult that we started and everybody killed themselves and uh there's a will release it then because it'll be like unseen footage, this is how it started. It would be black and white in the basements of war in new york, [/INST] my mother's </s><s>[INST] basement, let's explain so much. Okay, so I spoke to your aunt Brooke about objectivism and I ran he uh, he kind of argued, he highlighted difference between capitalism and anarchism as around the topic of violence and the that having government be the sort of the, the negative way to say it is like having a monopoly on violence, but basically being the arbiter of or the the people that making sure that violence doesn't get out of hand. That was 20 [/INST] 20 show that the government is great at that. </s><s>[INST] Well, what, what's okay without this [/INST] the same with a straight face making that argument good work you're on. </s><s>[INST] All right. Well, can you with a straight face argue ah, for the idea that and anarchism, violence would not get out of hand. [/INST] Sure. For one thing, uh, if your worst argument against and one of my little quotes is what are presented as the strongest arguments against anarchism are inevitably description of the stress quo. So the argument is under anarchism, you know, you have warlords, you know, killing people and then you'd have, you know, whoever's strongest gets to just take over a neighborhood. Well, we have that. Now we saw that the police are perfectly comfortable disarming the population and then when they try to protect themselves are punished, they were happy to stand down. You can't, you can only have that happen if you have a monopoly, if they're like, let's suppose you had television stations, right? And cbs said, you know what, We're not going to broadcast, cool, you don't broadcast, we're going to watch any of these other channels. So the problem with having monopoly is everyone has to be dependent on this issue. What's amazing about anarchism, which objectivist are, is they will argue that government is really, really bad at everything it does, and it touches therefore, it has to be in charge of the most important stuff. </s>

<s>[INST] Be a theater. Okay, hold on a second. I have to push back and this is me saying it and I'll, I'm sure I'll talk to Eric about this. But to say, you said Eric's done, was right on multiple things. I think Eric has a great deep insight about human nature and how societies work. And he says a lot of wise words on that world. But I think if we're talking about experts, you kind of have to prove you know, it's like Michael Jordan playing baseball. Like he's proved it many times that he can play basketball, but he's also got to prove that he can play baseball. And I would say the whole point of, I mean, of radical ideas is you're not, I mean, it's very hard to be sitting on a track record of, I mean, when you're swinging for the fences, always you're, there's not a track record to sit on. And like Max Tegmark is an example of somebody who has a huge track record of more like acceptable stuff, but he also keeps swinging for the fences in every other world. So he has that track record. With Eric, if you look at just the number of publications, all this stuff, he did really, he chose not to travel the academic route. So there's no proof of expertise except sort of an obvious linguistic demonstration of brilliance, [/INST] but that's not how physics works, right? That's not how physics works, right? There's a polite way to damn somebody as a scientist and say, he or she, they really know the history of physics very well. Like physicists always love it. Like Sean Carr always jokes about like, physicists should never talk about history of physics, but it's more than that. So Eric has certainly contributed in finance specifically and gauge theory in economics and inflation dynamics </s><s>[INST] and the non-cosmological, non-cosmological. Hold on a second, that's yet to be proven. [/INST] He has a lot of powerful interest, powerful interest. Well, gauge theory is calculus proven. I mean, he has a gauge model for currency exchanges between different nations that is explanatory. It's not, you know, is this something, in other words, it's a model </s><s>[INST] and it's used for pedagogical purposes. And it might be, okay. It's unique to him. I mean, to him and Pia. Yes, right. It might be a powerful model. It might be one that actually deserves huge amount of applause and celebration, but this not yet received that. And that's one of the things that Eric talks about it has not received the attention it deserves. But it has not yet received the attention it deserves. And so like the proven expertise thing, I mean, there's a lot of people that go to their grave without the recognition they deserve. And it's a tragedy. But the fact is, like you have to fight for that recognition. The tragedy happens for a reason. You can't just say, this person is obviously brilliant and therefore they deserve the credit in every single domain. It doesn't transfer immediately. There's nobody that's, well, at least I wouldn't argue, Eric is one of the special minds in our generation. But you still have to fight the fight of physics and prove it within the community. And I think the same applies in economics. You can't, as somebody that I've gone through the academic journey, just like you said, the peer review, all of those things, flawed as they are, that's the part of the process. You have to convince your peers, the people that are as obsessed for whatever the hell reason about that particular thing that you're working on. Yes, there's egos, yes, there's politics. It's a giant mess, but I think it's a beautiful mess through which you have to go through in order to reveal the power of your idea [/INST] to yourself and to the world. It's unique to him, I mean, to him and Pia. Yes, right. Let me use an example. So you know of James Clerk Maxwell and he invented the laws of electromagnetism, which is the first example of a unification principle ever displayed by the human mind in history, purely mathematics, unifying completely disparate phenomena. In one case, electricity, charges, static electricity, lightning, and the other magnets, bar magnets, currents, et cetera, unified them. You know what he did? I like to do a thought experiment. Imagine Twitter exists 1864, Maxwell's working away. And he goes, I have this wonderful idea with fluxions and inductive virtue and blah, blah, blah. And it revolves on this thing called an ether. And by the way, there are these little vortices and gears and the gears have these planetary things and they suck up vortices and the vortices determine the density of the electromagnetic potential. You'd be like, this guy's a fricking moron. And what would you do? Come on, honestly, you would say everything this guy does is wrong. I mean, he's got this idiotic idea. And it would be falsified a couple of decades later by Michelson and Morley. And in so doing, you would have thrown out a very beautiful baby with bathwater. Imagine Twitter, imagine a Twitch storm, at clerk Maxwell, one would get, it would be brutal, right? And to the detriment. And that might even set back history. Imagine Yang Mills doing the same thing. Churn Simons, a lot of these things are very fantastic. But why Lex? Why does Ed Witten? Why does Juan Maldisena? Let me give a good example. Juan, brilliant guy, I love him. He is the reason that Stephen Hawking conceded his black hole information paradox loss issue. What did he concede? Conceded based upon Maldisena's calculation in ADS, CFT, and five dimensional wormholes. But is any of that, first of all, we don't live in an ADS universe. Second of all, we don't know if wormholes are traversable, if they exist even. These are devices, Kip Thorne has popularized for movies. To say that this is something on which I will concede a bet. Now, obviously, Hawking was doing that for publicity. Why does Maldisena, and he's got a pretty high H index, pretty well-respected guy, IAS, love talking to him, brilliant guy. By the way, also had made use of Eric and Pia's work on gauge theory and economics originally. And one, I believe the breakthrough, I can't remember exactly what, but partially credit some of the work that he did, which appears there's a footnote to Pia Milani's thesis and some conversations with Eric, I think, in it. Anyway, getting back to that, why is there not the same skepticism? Is it because Maldisena, who's an eminent physicist, obviously, has published realistic work and done, what about Whitten? Whitten gets a pass, I mean. </s>

<s>[INST] But what is the thing you're, You know, when you sit back on a porch and just look out into the sunset. What are you most proud of from those 10 days? [/INST] I think the first class functions shines I think especially since job I didn't have it and it was somewhat unusual scheme </s><s>[INST] made it in somehow, uh, at the end of the [/INST] day in spirit people complain because scheme has, you know, minimalism. It has, you know, six or seven special forms. It has hygienic macros, it has call cc. It has sort of beautiful, complete um, set of forms to make the lambda calculus pleasant to use in practice. Um, and javascript is, you know, kind of uh, multi paradigm or shambolic just </s><s>[INST] in a small tangent. You mentioned marc Andresen, it sounds like and Bill Joy but staying on mark, it sounds like he had an impact on you in that. He sort of believed in what you were doing there. Can you, can you talk about like what role Mark had in your [/INST] life? Yeah. We would meet at the, the pencil creamery in downtown Palo Alto and Mark was just fresh out of grad school or whatever he was doing and he was big dude and he got federal later. Uh, he had hair, he, he would order giant milkshakes and burgers and we would meet there and brainstorm about what to do. And it was very direct because we didn't have much time. The sort of, we didn't talk about it. The implication was Microsoft's coming after us. Mark was saying things boldly Predappio like Netscape plus java kills windows. Right? This was ambitious makeup browser parable. It becomes the new runtime for programs, you know, it's the meadow assets, replacement. Os um, </s><s>[INST] but he still saw value in [/INST] javascript. Yes. Even though he was saying that in java was the big name, hence the trademark license. Uh, he saw javascript is important and he even thought, what if we got, I told this in other interviews, I can say he thought, what if we had my friend kipp Hickman who had been at Netscape from the beginning and who was a kernel hacker S. G I when I joined, he started writing his own JBM before we Consummated the Sun deal and got our hands on their code. And the java compiler, java C which Arthur Van Hoff had written very nice code was all written in java it was self hosted or so called Bootstrap and so we could use that as soon as Kipps, Java VM could run the bike code from the Sun. Uh, JBM running the self hosting compiler to admit the bike. Oh, so once we could bootstrap into kIPs fee and we wouldn't need Sun. And Mark was like, well maybe we can just, you know, ditch son of Java VM or if your javascript VM we now we need graphics. So Mark was thinking far ahead because he knew you could do things with html and images, but at some point you really want the same demographics or like even S G I had already started its downfall because the first four to be less I team there had gone off to do three DfX and all these other companies that made the graphics card in your Pc right. Doom was big and quake. And so you were, we were all playing quake. I was told I was terrible, but why not put that graphics capability on the web? And in fact it finally happened uh at Mozilla Firefox area with Vlad Vukicevic taking open Gl es and reflecting in his web Gl but open Gl es is the mobile version of Open Gl which is a standard based on S G I G L. So this whole lineage of graphics libraries or really graphics languages for the, what became the Gpu? And Mark was thinking ahead. It's like we need graphics too. And I thought, okay, I try to get somebody I knew it S D. I. But he's a grad student at M. I. T. Was studying under Barbara list cough. He laughed when he heard about this later. Andrew Myers. He's a Cornell long time and he's a full professor. And Mark said, great, we'll get him and I'm not sure he's going to come, we'll throw money stock options. We never did it. And they did the sun deal. So keep nobly put aside his own JBM. And we used the sun JBM. So that was an ambitious period. And Mark was very generative because he was pushing hard. He was ambitious and he wanted to have Netscape possibly be in control of the ball. </s><s>[INST] Maybe you can speak to the this dance of uh Netscape versus internet Explorer. You've uh you've thrown some loving words towards Microsoft throughout this conversation, but that's a theme with I mean steve jobs had a similar sort of commentary from a big sort of philosophical principle perspective. Can you comment on like the approach that Microsoft has taken with Internet Explorer from ie one to Edge today. Is there something that you see as valuable that they're doing in the in the occasional copying and that that kind of stuff? Or is it um is the world worse off because an Explorer exists? [/INST] So I'm going to segment this into historical errors because I think Microsoft today with is quite a different company and what they're doing with edges different. But back then um Gates, you know, aggressive character, not really original in my view, uh not an originator. Steve jobs famously said once he doesn't have any taste and I don't mean this in a small way, there's no taste. You can see this. Apple at the time had beautiful typography and your ligatures and turning and the fonts look great. And Windows had this sort of ugly system font that was carefully aligned with pixels. So it I didn't get why, </s>

<s>[INST] There's quite a lot that goes into it. Or just. So there's a lot of problem solving. Like you've already said for GPT four in the blog post and in general, there's already kind of a maturity that's happening on some of these steps, like being able to predict before doing the full training of how the model will behave. [/INST] Isn't that so remarkable by the way, that there's like, you know, there's like a law of science that lets you predict for these inputs, here's what's gonna come out the other end. </s><s>[INST] Like here's the level of intelligence you can expect. Is it close to a science or is it still, because you said the word law and science, which are very ambitious terms. Close to, I suppose. Close to, right. [/INST] Be accurate, yes. I'll say it's way more scientific than I ever would have dared to imagine. </s><s>[INST] So you can really know the peculiar characteristics of the fully trained system [/INST] from just a little bit of training. You know, like any new branch of science, there's we're gonna discover new things that don't fit the data and have to come up with better explanations. And, you know, that is the ongoing process of discovering science. But with what we know now, even what we had in that GPT four blog post, like, I think we should all just like be in awe of how amazing it is that we can even predict </s><s>[INST] to this current level. Yeah, you can look at a one year old baby and predict how it's going to do on the SATs. I don't know, seemingly an equivalent one, but because here we can actually in detail introspect, various aspects of the system you can predict. That said, just to jump around, you said the language model that is GPT four, it learns in quotes, something. In terms of science and art and so on, is there within open AI, within like folks like yourself and LES discover and the engineers, a deeper and deeper understanding of what that something is, or is it still a kind of beautiful, [/INST] magical mystery for the system you can... Well, there's all these different evals that we could talk about. And... What's an eval? Oh, like how we measure a model as we're training it after we've trained it and say like, you know, </s><s>[INST] how good is this at some set of tasks? And also just a small tangent, thank you for sort of open sourcing the evaluation process. [/INST] Yeah. Evaluation process, yeah. I think that'll be really helpful. But the one that really matters is, you know, we pour all of this effort and money and time into this thing. And then what it comes out with, like how useful is that to people? How much delight does that bring people? How much does that help them create a much better world, new science, new products, new services, whatever. And that's the one that matters. And understanding for a particular set of inputs, like how much value and utility to provide to people, I think we are understanding that better. Do we understand everything about why the model does one thing and not one other thing? Certainly not always. But I would say we are pushing back like the fog of war more and more. And we are, you know, it took a lot of understanding </s><s>[INST] to make GPT-4, for example. But I'm not even sure we can ever fully understand. Like you said, you would understand by asking it questions essentially, because it's compressing all of the web, like a huge sloth of the web into a small number of parameters, into one organized black box that is human wisdom. What is that? [/INST] Human knowledge, let's say. </s><s>[INST] Human knowledge, human knowledge. It's a good difference. Is there a difference between knowledge? There's other facts and there's wisdom. And I feel like GPT-4 can be also full of wisdom. [/INST] What's the leap from facts to wisdom? Fast to wisdom. You know, a funny thing about the way we're training these models is I suspect too much of the like processing power for lack of a better word is going into using the models as a database instead of using the model as a reasoning engine. The thing that's really amazing about the system is that it, for some definition of reasoning, and we could of course quibble about it and there's plenty for which definitions this wouldn't be accurate. But for some definition it can do some kind of reasoning. And, you know, maybe like the scholars and the experts and like the armchair quarterbacks on Twitter would say, no, it can't. You're misusing the word, you know, whatever, whatever. But I think most people who have used the system would say, okay, it's doing something in this direction. And I think that's remarkable. And the thing that's most exciting and somehow out of ingesting human knowledge, it's coming up with this reasoning capability. However, we're gonna talk about that. Now, in some senses, I think that will be additive to human wisdom. And in some other senses, you can use GPT-4 for all kinds of things and say that appears </s><s>[INST] that there's no wisdom in here whatsoever. Yeah, at least in interactions with humans, it seems to possess wisdom, especially when there's a continuous interaction of multiple prompts. So I think what, on the chat GPT side, it says the dialogue format makes it possible for chat GPT to answer follow-up questions, admit its mistakes, challenge incorrect premises and reject inappropriate requests. [/INST] But also there's a feeling like it's struggling with ideas. Yeah, it's always tempting to anthropomorphize </s><s>[INST] this stuff too much, but I also feel that way. Maybe I'll take a small tangent towards Jordan Peterson who posted on Twitter this kind of political question. Everyone has a different question they wanna ask chat GPT first, right? [/INST] Like the different directions you wanna try the dark thing. It's somehow says a lot about people. </s>

<s>[INST] So the intelligent constructions of physical systems that would embody the intelligence versus container within the computation. [/INST] Right, and there's a brain centrism that assumes our intelligence is centered in our brain. And in endless ways in this conversation, we've been talking about molecular intelligence. Our molecular systems do a deep kind of artificial intelligence, all the things you think of as artificial intelligence does in representing knowledge, storing knowledge, searching over knowledge, adapting to knowledge our molecular systems do. But the output isn't just a thought, it's us. It's the evolution of us. And that's, the real horizon to come is now embodying AI of not just a processor and a robot, but building systems that really can grow and evolve. </s><s>[INST] So we've been speaking about this boundary between bits and atoms. So let me ask you about one of the big mysteries of consciousness. [/INST] Do you think it comes from somewhere between that boundary? I won't name names, but if you know who I'm talking about, it's probably clear. I once did a drive, in fact, up to the Mussolini-era villa outside Torino in the early days of what became quantum computing with a famous person who thinks about quantum mechanics and consciousness. And we had the most infuriating conversation that went roughly along the lines of consciousness is weird, quantum mechanics is weird, therefore quantum mechanics explains consciousness. </s><s>[INST] That was roughly the logical process. [/INST] And you're not satisfied with that process? No, and I say that very precisely in the following sense. I was a program manager somewhat by accident in a DARPA program on quantum biology. And so biology trivially uses quantum mechanics and that were made out of atoms, but the distinction is in quantum computing, quantum information, you need quantum coherence. And there's a lot of muddled thinking about like collapse of the wave function and claims of quantum computing that garbles just quantum coherence that you can think of it as a wave that has very special properties, but these wave-like properties. And so there's a small set of places where biology uses quantum mechanics in that deeper sense. One is how light is converted to energy in photo systems. It looks like one is olfaction, how your nose is able to tell different smells. Probably one has to do with how birds navigate, how they sense magnetic fields. That involves a coupling between a very weak energy with a magnetic field, coupling into chemical reactions. And there's a beautiful system. Standard in chemistry is magnetic fields like this can influence chemistry, but there are biological circuits that are carefully balanced with two pathways that become unbalanced with magnetic fields. So each of these areas are expensive for biology. It has to consume resources to use quantum mechanics in this way. So again, those are places where we know there's quantum mechanics and biology. In cognition, there's just no evidence. There's no evidence of anything quantum mechanical going on in how cognition works. Consciousness. I'm saying cognition, I'm not saying consciousness, but to get from cognition to consciousness... So McCulloch and Pitts made a model of neurons. That led to perceptrons that then, through a couple of boom busts, led to deep learning. One of the interesting things about that sequence is it diverged off so deep neural networks used in machine learning, diverged from trying to understand how the brain works. What makes them work, what's emerged is, it's a really interesting story. This may be too much of a technical detail, but it has to do with function approximation. We talked about exponentials. A deep network needs an exponentially larger, shallow network to do the same function. And that exponential is what gives the power to deep networks. But what's interesting is the sort of lessons about building these deep architectures and how to train them have really interesting echoes to how brains work. And there's an interesting conversation that's sort of coming back of neuroscientists looking over the shoulder of people training these deep networks, seeing interesting echoes for how the brain works, interesting parallels with it. And so I didn't say consciousness, I just said cognition, but I don't know any experimental evidence that points to anything in neurobiology that says we need quantum mechanics. And I view the question about whether a large language model is conscious as silly in that biology is full of hacks and it works, there's no evidence we have that there's anything deeper going on </s><s>[INST] than just this sort of stacking up of hacks in the brain. And somehow consciousness is one of the hacks [/INST] or an emergent property of the hacks. Absolutely. And just numerically, I said big computations now have the degrees of freedom of the brain. And they're showing a lot of the phenomenology of what we think is properties of what a brain can do. </s><s>[INST] And I don't see any reason to invoke anything else. That makes you wonder what kind of beautiful stuff digital fabrication will create. If biology created a few hacks on top of which consciousness and cognition and some of the things we love about human beings was created. It makes you wonder what kind of beauty in the complexity [/INST] can create as a digital fabrication. There's an early peak at that, which is there's a misleading term, which is generative design. Generative design is where you don't tell a computer how to design something, you tell the computer what you want it to do. That doesn't work, that only works in limited subdomains. You can't do really complex functionality that way. The one place it's matured though is topology optimization for structure. So let's say you wanted to make a bicycle or a table. You describe the loads on it and it figures out how to design it. And what it makes are beautiful organic looking things. These are things that look like they grew in a forest and they look like they grew in a forest because that's sort of exactly what they are, that they're solving the ways of how you handle loads in the same way biology does. And so you get things that look like trees and shells and all of that. And so that's a peak at this transition to from we design to we teach the machines how to design. </s>

<s>[INST] that's that's why I played with stock fisher bunch. So I I got um I built for myself a bunch of different chess experiments recently, I had to do with the chess playing bot. But I also built the infinite chessboard. Where was stock fish was playing like an infinite number of chess games. And one of the parameters that was interesting to play with is how long it gets to think about a move and how that affects the rating of the thing [/INST] I did that a little bit. </s><s>[INST] So that that's a tricky one, I'm sure people know how to do that well, but it's tricky, it's not trivial to understand, like what there must be a good formula for it, but it's also interesting to think about like a controlled number of blunders, but it's probably better. The control number of blunders is not a good way to build a blot. [/INST] Yeah, it's for </s><s>[INST] training purposes, at least the [/INST] time the time per move is probably better, But the craziest thing is I did that a couple of my of my deaths were helping me with the build like, uh, I'm scaling my courses into a better chest learning platform essentially. We've done a lot of different experiments with stock fish, which I'm even happy to get into here. Um, and uh, stock fish making moves in .1 millisecond plays better than a human, which </s><s>[INST] is disgusting [/INST] and disturbing frankly, because that's crazy. Like you can't react to a car stopping in front of you anywhere near that fast. And yeah, </s><s>[INST] so the reason I was interested in that is because when you have a very large chessboard, you have tens of thousands, hundreds of thousands of games going on at the same time. You have the, the, like, think of the minimum amount of thinking to remove that you can allow for. And it does, it does, it's damn good, at least to my, I basically at the lowest possible setting you can give it. So yeah, it's incredible. It's incredible what bots are able to do and now it's as far as I know, as primarily machine learning based stock fish, stock fish moved to machine learning completely not doing search as far as I know. [/INST] Yeah, this is where you you lose me a bit. But the move, discovery and evaluation is what's been changing in the way starfish works. So it's discovering of moves and then the way it looks forward and then evaluates positions that has changed. But Matthew Sadler, who wrote the book, game changer about Alpha </s><s>[INST] Zero. Yeah, [/INST] that that he explains it significantly better, but that that's the way I think it works. </s><s>[INST] What did that make you feel when you first saw α zero play? [/INST] I was excited. I I didn't have any any sort of existential thoughts. I enjoyed watching it completely destroy openings that people thought were good. Uh And it it that experiment though has does have some caveat in the sense that I think Alpha Zero was playing with a full tank of servers and I think stock fish wasn't, which I think is what one of the things people point out, they weren't </s><s>[INST] playing since then they were able to demonstrate much [/INST] less </s><s>[INST] uh But also α zero stopped developing. Um they stopped developing it, [/INST] Yes, </s><s>[INST] which which sucks, but again from their perspective from deepmind perspective, it's like, alright, well, we took on this really, Turkey game did something honestly incredible. [/INST] They won, they won, I mean, </s><s>[INST] not just one, I mean they did it without any human supervision. So like without any training on human expert games, so only through self play, which I mean that's what learning is about, it's like that's what you think of as a child, a human child, a toddler learning from, [/INST] you </s><s>[INST] know, somewhat nothing and becoming a capable human. That's that's what we think of when we think about intelligence. So the fact that's able to play itself and become the best player in the world that at the game of go and all kinds of games is just incredible. And obviously that inspired the modern stock fish. Yeah. Now to do the same, all the same kind of self play methods. Uh, somebody on Reddit ask a pretty interesting question. [/INST] I </s><s>[INST] don't know if you have an interesting answer to it. What makes the chess move, quote unquote human as you are someone? Oh, this is to me, but it's mentioned you in third person as you are someone working in ai this idea of humanness would seem incredibly interesting. Yes sir, it is, especially since most cheat detection relies on humanness as a way to detect cheaters. I think since Children being born right now will have the advantage of engine training their whole life, they will start to see the game the way an engine does. Will a person be considered a cheater if they play like an engine, there also seems to be a discrepancy, especially with levy about who can play, what appears to be a non human move. He often says things like quote, if you were a normal player, normal in quotes, if you were a normal player, I would think you were cheating. But since it's magnus, I don't doubt magnus is great, but if humanness is our benchmark, what is the elo rating, where your moves can start to look like an engine without critique? There's a bunch of questions in there, [/INST] they combined two of my quotes into one. So one thing that I like to say is sometimes in a chess game moves, look or an opening look so ridiculous that if a viewer played it, I would make fun of them or slap them. That's always there. If I was your chest coach, I smack you. But when magnus plays, it was, oh wow, you know, it's a very different. So they mixed up this quote with another quote, which is uh you know, if, if I, if I'll be explaining something and I'll say, oh and here, the engine says you should play like this. If one of your opponents plays like this, report them for cheating, they fuse two quotes. They said, oh human can't play like that. But because it's an engine move, but because magnus is playing, so they mixed kind of two things </s>

<s>[INST] That's true. [/INST] And I think that it demands his argument was um there's a tremendous demand on working memory to work out theorems in math and to keep a number of plates spinning so to speak mentally and run back and forth between them, updating them in physics Jack said and I mean I think this makes sense to me too that there's a reliance on working memory but an increased reliance on some sort of deep deep memory and deep memory stores probably stuff that's moved out of the hippocampus and for draining into the cortex and is um more some episodic and declarative stuff, but really so you're you're pulling from your library basically, it's not all RAM, it's not all working memory and then in biology that and physicists tend to have very active careers into their thirties and forties and fifties and so forth. Um sometimes later and that in biology you see careers that have a much longer arc kind of these protracted careers, often people still in their sixties and seventies doing doing really terrific work, not always doing it with their own hands because the people in the labs are doing them of course, but um and that work does tend to rely on insights gained from having a very deep knowledge base where you can remember a paper and or maybe a figure in a paper you could go look it up if you wanted to but it's very different than the working memory of the mathematician and so when you're talking about coding or being in that tunnel of thought and trying to iterate and keeping a lot of plates spinning. It speaks directly to working memory. My lab hasn't done too much of that working memory. But we are pushing working memory when we have people do things like these simple lights out tasks while they're under. We can increase the cognitive load by increasing the level of autonomic arousal to the point where they start doing less well. And you know, everyone has a cliff. This is what's kind of fun. We've had, you know, seal team operators come to the lab, we have people from other units in the military very, you know, we've had a range of of intellects and backgrounds and all sorts of things and everyone has a cliff and those cliffs uh sometimes show up as a function of the demands of speed of processing or how many things you need to keep online. I mean we're all limited at some point in the number of things we can keep online. So what you're describing is very interesting because I think it has to do with how narrow or broad the information set is because and I don't program, I'm not an active programmer. So this is a regime I don't really fully know. So I don't want to comment about it in any way. Uh you know, doesn't suggest that, but I think that what you're talking about is top down control. So this is prefrontal cortex keeping every bit of reflexive circuitry at bay. The one that makes you want to get up and use the restroom the one that makes you want to check your phone all of that. But also running these interior thalamus to prefrontal cortex loops which we know are very important for working memory. </s><s>[INST] Yeah. Let me try to think that it's a little bit so reducing the process of thinking to working memory access. It's tricky. He's probably ultimately correct. But if I were to say some of the most challenging things that an engineer has to do and the scientists, scientific thinker, I would say it's kind of depressing to think that we do that best in our twenties but is uh this kind of first principles thinking step of saying [/INST] you you're </s><s>[INST] accessing the things that you know and then saying, well let me how do I do this differently than I've done it before this this weird like stepping back like is this right? Let's try it this other way that that's the most mentally taxing step is like you've gotten quite good at this particular pattern of how you solve this particular problem. So there's a there's a pattern recognition first, you're like okay, I know how to build the thing that solves this particular problem in programming, say. And then the question is but can I do it much better and I don't know if that's I don't know what the hell that is. I don't know if that's accessing working memory that's that's almost access, maybe it is accessing memory in a sense. It's trying to find similar patterns in a totally different place that could be projected onto this, but you're it's you're not quarrying a fact. You're currying like functional things like yes, patterns. I mean, you're running out your [/INST] your testing algorithms. Yeah, right? You're testing algorithms. Ice. So I want to just um because I know uh some of the people listening to this and you how have basis and you know, scientific training and have scientific training. So I want to be clear. I think we can be correct about some things like The role of working memory in these kinds of processes without being exhaustive. We're not saying that the only thing, you know, we could be correct but not assume that that's the only thing involved. And I mean, neuroscience, let's face it is still in its infancy. I mean, we probably know 1% of what there is to know about the brain. Um you know, we've learned so much and yet there may be global states that underlie this that make prefrontal circuitry work differently than it would in a in a different regime even time of day. I mean there's a lot of mysteries about this, but so I just want to make sure that we we sort of are we're aiming for precision and accuracy, but but we're not gonna we're not gonna be exhausted. So there's a difference there and I think uh you know, sometimes in the vastness of the internet that gets forgotten. Um So the other is that um you know we think about um you know we think about these operations uh at you know really focused keeping a lot of things online. But what you're describing is actually um it it speaks to the very real possibility. Probably the with certainty there's another element to all this which is when you're trying out lots of things in particular, lots of different algorithms. You don't want to be in a in a state of very high autonomic arousal. That's not what you want because the higher level of autonomic arousal and stress in the system, the more rigidly you're going to analyze space and time and what you're talking about is playing with spacetime dimensionality. And I want to be very clear. I mean I'm the son of a physicist. I am not a physicist. When I talk about space and time, I'm literally talking about visual space and </s>

<s>[INST] the hell is that about the human spirit? That's that's [/INST] that's the thing about Antarctica is it brings it out of you. For example, I read another one recently called Moss ins will Douglas Mawson. He was an Australian. He was on the, one of the first shackled frost robert frost expeditions. He leads an expedition down to the south. Him and a partner there, leading explorations 19 oh eight, something like that. They're going around Antarctica um with dog teams. And one of the, what happens is they keep going over these snow bridges where there's a crevice, but it's covered in snow. And so one of the lead driver, the dogs go over and they plummet and that sled takes with it. So the guy survives, but that sled takes all their food, half the dogs, their stove, the camping tent, the tent specifically designed for the snow, everything. And there are hundreds of miles away from base camp. He and this guy, I have to make it back there in time before the ship comes to come get them on an agreed upon date. And he makes it. But the guy he was with, he dies and it's a crazy story. They first of all, we have to eat the dogs, A really creepy part of antarctic exploration as everyone ends up eating dogs at different points. Um and part of the theory, which is so crazy is that the guy he was with was dying because they were eating dog liver and dog liver has a lot of vitamin E, which if you eat too much of it can give you like a poisoning. And so uh Mawson by trying to help his friend was giving him more, </s><s>[INST] all the things that kills, you know, [/INST] his dog liver, his friend ends up dying, have a horrific heart attack. All of that, Mawson crawls back hundreds of miles away, makes it back to base camp hours after the ship leaves. And two guys are, a couple of guys stayed behind for him and he basically has to recuperate for like six months before you can even walk again. But it's like you were saying about the human spirit, it's like Antarctica brings that out of people or almonds on. The guy who made it to the South Pole robert Ahmanson, oh my God, like this guy trained his whole life in the ice from Norway to make it to the South Pole and he beat robert Frost, the british guy with all this money and all these. I could go on this forever. I'm </s><s>[INST] obsessed with. Well, first of all, I'm gonna, I'm gonna take this part of the podcast. I'm gonna set it to music. Gonna listen to it because I've been whining and bitching about running 48 miles of Goggins this next weekend and this, this is gonna be so easy. I'm just gonna listen to this over and over in my head, you're going [/INST] to be one. He was obsessed with Shackleton. He talks about, he </s><s>[INST] uses, I was going to ask you about the use as an example of that is as an example of what mars colonization would be like. He's [/INST] right. No, that Antarctica is as close to, you can simulate that. Um the antarctic goes as close to what you can simulate what it would get. That that nat geo series on mars, I'm not sure if you watched it. It's incredible. Ellen is actually in it. Um and it's kind of, and it's like they get there, everything goes wrong, somebody dies like it's horrible. They can't find any water, it's not working. So what </s><s>[INST] is it? Is it like simulating the experience of what it be [/INST] like? So it's like a dock, you series where the fictionalized part is the astronauts on mars, but then they're interviewing people like Elon musk and others who were the ones who paved the way to get to MArs. This is a really interesting concept. I think it's on netflix And yeah, I agree with him 100%. Which is that the first guys to make like for example, Robert Frost who uh went to Australia. Well, sorry to Antarctica. The british explorer who was beaten to the South Pole three weeks by robert ahmanson, he died on the way back and the reason why because he wasn't well prepared. He was arrogant, he didn't have the proper amounts of supplies. His team had terrible morale. Antarctica was a brutal place if you funk up one time you die and it's like you and this is what you read a lot about, which is the reason why such heroic characters like Shackleton Shine is a lot of people died. Like there were some people who got frozen in the eye. I mean man, this again also came to the north Exploration. I read a lot about like the exploration of the North Pole and the same thing. These un extraordinary men Take people out into the ice and get frozen out there for years and ship goes so bad. They end up eating each other. They all die. There's a famous I want, I'm forgetting his name. The British Franklin expedition where they went searching for them for like 20 years and they eventually came across a group of Inuit who were like, oh yeah, we saw some weird white men here like 15 years ago and they find their bones and there's like saw marks which show that they were eating each other. </s><s>[INST] So history remembers the ones who didn't eat each other. [/INST] We were well yeah, we remember the ones who made it, but there are </s>

<s>[INST] Well, it could be, you know, you could have the Ernest Becker theory of denial of death where at the core of that the warm of the core, as young said, is the fear of death. So were the longing for the perfect thing. It has to do with sort of becoming immortal is um reaching beyond the absurdity, the cruelty of life that all things come to an end for, for no particularly good reason whatsoever. One we can rationally explain. [/INST] I know, you know, I wonder about that all the time. Like I know obviously there's that idea from Becker and throughout philosophy and the tale of Gilgamesh about the idea that the thing we're longing for most of all is immortality, </s><s>[INST] but [/INST] I feel like it's not only that, I think it's more so or also let's say, um, a longing for the lions to lay down with the lambs. Finally, you know, for like the, the fundamental calculus of the universe to be different where life doesn't have to eat life in order to survive. And yeah, just a completely different situation, immortality would not solve. </s><s>[INST] I wonder that could be a very kind of modern thing because surely so much of human history is defined by violence and glorified violence. That doesn't give inklings of this lions and the lambs so much. Yeah, I [/INST] mean, I know all the other stuff is in the bible, </s><s>[INST] there's other stuff in the bible and the bible and that particular aspect doesn't necessarily reveal the fundamental motivation of human nature. That could be deeper stuff, you know? But yeah, that is a beautiful, that is a beautiful picture. But is it just about humans or is it all about about all of life? And you have to think about what is the perfect world look like? It's not just the lines and lambs laying together is you know, how many lines and how many [/INST] lambs and </s><s>[INST] uh you know what, having just had a few very technical conversations about marxian economics versus kensi and economics versus neoclassical economics. [/INST] What does the </s><s>[INST] economic and the government system look like for the lines and the labs that were longing [/INST] for? So </s><s>[INST] that then you start to build society on top of all those things and but you still, you return to this, what is it? What are we longing for and what's the role of love and that what's the role of that sad melancholy feeling? The feeling of loneliness is the feeling of loneliness fundamental to the human condition, Like? Are we always striving sort of channel that feeling of loneliness um to connect with others? Like we want that feeling of loneliness, otherwise we wouldn't be connecting. Is that fundamental? That feeling like you're alone in this, even when you're with other people sort of alone together, you're born alone, you die alone. Maybe loneliness is fundamental. [/INST] I think the longing for Union is fundamental. It's just that it looks so different for different people, you </s><s>[INST] know? Yeah. [/INST] Coming back to where we were, what we were talking about at the beginning, you know, Union looks incredibly social for a lot of people and hardly social at all for others, but everybody needs some version of union. </s><s>[INST] Yeah. People have been telling me recently about polyamory and all those kinds of things. Some having probably grown up in a certain part of the world. I'm very, I think I'm very monogamy centric, not in a judgmental way. Just for me, what makes me happy is one person feel for my whole life, basically just dedication. Because I've just seen through relationships with people and objects in my life, the longer we stay together, the deeper the tie. So that's just the empirical thing. And yes, that probably it's a personalized thing that's just true for me, it could be very different for others. Maybe it's connected to the introverted thing? Maybe not Who knows before before, because you mentioned songs, sad songs? What um what are we talking about? What's a good what what what song do you remember last crying two? [/INST] Oh gosh, well, I mean that's you know, I I </s><s>[INST] literally [/INST] literally dedicated my book to Leonard Cohen, he's played such a huge role in my life, I love him, I love him. Um and I loved him with this crazy love that I've never been able to understand for decades. I think I understand it a little better now, </s><s>[INST] but you guys, so you're better friends with him than me? I'm so um does it make you, is it the the musician or the human too? Because the human is a is a tortured soul in a way, [/INST] I'd say it's the musician, it's the musician and I actually was thinking about this the other day. I mean obviously he's not alive anymore, but I was kind of running the thought experiment, you know, if he were alive still and I had the chance to meet him </s><s>[INST] in person, [/INST] would I want to do that? And I'm not really sure that I would because he represents for me symbolically everything. Well everything else in the sentence right there. And so and I think that's okay, you know, I think people can can express something through their art that they might or might not express if you were just like hanging out with them and having a coffee and I'm happy to know him that way. </s>

<s>[INST] that context puts all the different modalities on the same level ground provide the context. So maybe on that point. Uh So there's this task which may not seem trivial of token. Izing the data of converting the data into pieces into basic atomic elements that then could uh cross modality somehow. So what's tokenization? How do you tokenize text? How do you tokenize images? How do you tokenize games and actions and robotics tasks? [/INST] Yeah, that's a great question. So tokenization is the entry point to actually make all the data look like a sequence because tokens then are just kind of this little puzzle pieces. We break down anything into these puzzle pieces and then we just model what what's this puzzle look like? Right, when you make it you know, laid down in the line, so to speak in a sequence. So in gatto um the text, there's a lot of work you do recognize text usually by looking at common commonly used sub strings. Right? So there's you know I N. G in english is a very common sub string. So that becomes a token. Um there's quite well studied problem on organizing text. Text and gotto just use the standard techniques that have been developed from many years, even starting from n gram models in the 19 fifties and so on. </s><s>[INST] Just for context, how many tokens, like what order magnitude number of tokens is required for a word? [/INST] Yeah. </s><s>[INST] Well what are we talking about? [/INST] Yeah, for a word in english. Right. I mean every language is very different from the current level or granularity of organization generally means is maybe 2-5. I mean I don't know the statistics exactly. But to give you an idea we don't tokenize at the level of letters then it would probably be like, I don't know what the average length of a word is in English but that would be, you know, the minimum set of tokens you could use. It </s><s>[INST] was bigger than letters, smaller than words. [/INST] Yes. Yes. And you could think of very very common words like the I mean that would be a single token but very quickly you're talking 2344 tokens or have </s><s>[INST] you ever tried to tokenize emojis? [/INST] Emojis are actually just um sequences of letters. So </s><s>[INST] maybe to you. But to me they mean so much more, [/INST] Yeah, you can render the emoji but you you might if you actually just </s><s>[INST] this is a philosophical question is emojis an image or a. Text. [/INST] The way we do these things. These things is they're actually mapped to small sequences of characters. So you can actually play with these models and input emojis. It will output emojis back which is actually quite a fun exercise. You probably can find other tweets about this out there. But yeah, so anyways, text there's like it's very clear how this is done. And then in Gatto what we did for images is we map images to essentially we compressed images, so to speak into something that looks more like um less like every pixel with every intensity. That would mean we have a very long sequence, right? Like if we were talking about 100 by 200 pixel images, that would make the sequences far too long. So what was done there is you just use a technique that essentially compresses an image Into maybe 16 x 16 patches of pixels. And then that is map again, tokenize you just essentially quantized this space into a special word that actually maps to these little sequence of pixels. And then you put the pixels together in some raster order and then that's how you get out or in or in the image that you're process, </s><s>[INST] there's no semantic aspect to that. So you're doing some kind of you don't need to understand anything about the image in order to tokenize it currently. [/INST] No, you're only using this notion of compression. So you're trying to find common, it's like J. P. D. Or all these algorithms, it's actually very similar at the organization level all we're doing is finding common patterns and then making sure in a lossy way we compress these images given the statistics of the images that are containing all the data we deal with. </s><s>[INST] Although you could probably argue that jpeg does have some understanding of images like because visual information maybe color compressing based crudely based on color does capture some something important about an image that's about its meaning, not just about some statistics. Yeah, [/INST] I mean jp as I said, it's very the algorithms look actually very similar to the use the the design transforming J P G. Um the the approach we usually do in machine learning when we deal with images and we do this quantization step is a bit more data driven. So rather than have some sort of food, air bases for how you know, frequencies appear in natural in the natural world. We actually just use the statistics of the images and then quantized them based on the statistics, much like you doing words right. So common sub sub strings are allocated a token. Um and images is very similar but there's no connection, the token space if you think of like the tokens are an integer and in the end of the day so now like we work on maybe we have about let's say I don't know the exact numbers but let's say 10,000 tokens for texts. Right? Certainly more than characters because we have groups of characters and so on. So from 1 to 10,000 those are representing all the language and the words we'll see. And then images occupy the next set of interest so they're completely independent. Right? So from 10001-20000, those are the tokens that represent these other modality images. And that is an interesting aspect. That makes it orthogonal. So what connects these concepts is the data? Right? Once you have a data set for instance that captions images that tells you, oh this is someone playing a frisbee on on a on a green field. Now the model will need to predict the tokens from the text Greenfield to then the pixels and that will start making the connections between the tokens. So these connections happen as the algorithm learns. And then the last if we think of these integers, the first few are words. The next few images in gatto We also allocated the highest order of integers. Two actions, right? Which we described ties and actions are very diverse. Right? In Atari there's I don't know if 17 discrete actions in robotics actions might be torques and forces that we apply. So we just use kind of similar ideas to compress these actions into tokens. And then we just that's how we map Now all the space to these sequence of integers but they occupy different space. And what connects them is then the learning algorithm that's where the magic happens. </s>

<s>[INST] The way you die as a philosophical question. But it's also a sociological and psychological question because it seems like every single person who's passed away over the past couple of years, kind of the first question that comes to mind just was a covid, not, not just because you're trying to be political, but just in [/INST] your mind. No, I think there's a psychological reason for this, right? So, you know, we have, we spent the better part of at least a half century in the United States, not worried too much about infectious diseases. And we, the notion was we essentially conquered them. It was something that happens in far away places to other people. And, and that's true for much of the developed world life expectancy. We're going up for decades and decades. And for the first time in living memory, we have a disease that can kill us. I mean, I think we're effectively evolved to fear that like the panic centers of our brain, the lizard part of our brain takes over and our central focus has been avoiding this one risk. Um, and so it's not surprising that people when they're filling out death certificates or thinking about what what what led to the death. This most salient thing that's in the front of everyone's brain would would jump to the top </s><s>[INST] and we can't ignore this very deep psychological thing. When we consider what people say on the internet, what people say to each other, what people write in scientific papers with everything, it feels like when Covid has been um has been brought into this world, everything changed in the way people feel about each other, just the way they communicate with each other. I think the the level of emotion involved. I think it in many people, it brought out the worst in them for sometimes short periods of time. And sometimes it was always therapeutic, like you were waiting to get out like the darkest parts of you just to say if you're angry at something in this world, I'm going to say it now. And um I think that's probably talking to some deep primal thing that um fear we have for formalities of all different kinds. And then when that fear is aroused in all the deepest emotions, it's like a Freudian psychotherapy session. But across the [/INST] world it's something that psychologists are gonna have a field day with for generations trying to understand. I mean, I think, I think what you say is right, but piled on top of that is also this sort of this impetus to empathy to empathize compassion toward others, essentially militarized. Right? So I'm protecting you. Bye bye. Some actions and uh, those actions, if I, if I don't do them, if you don't do them. Well, that must mean you don't, you hate me. Um, it's, it's created this like social attention that I've never seen before. And, and we, we started started, we started, we looked at each other as if we were just simply sources of, of germs rather than people to get to know people to enjoy, people to get, you know, sort of to learn from. Um, it colored basically almost every human interaction for every, every human on the planet. </s><s>[INST] Yeah. The basic common humanity. It's like you can wear a mask, you can stand far away. But the love you have for each other when you look into each other's eyes, that was dissipating and by region to have experienced having traveled quite a bit throughout this time. It, um, it was really sad. Even people that are really close together just the way they stood, the way they looked at each other and it made me feel for a moment, um, that the fabric that connects all of us is more fragile than I thought. [/INST] I mean if you walk down the street or if you ever, if you did this during Covid, I'm sure you had this experience where you walk down the street if you're not wearing a mask or even if you are, people will jump off the off the sidewalk that you walk past them as if you're poisoned, even though the data are, that Covid spreads, you know, indifferently outdoors if at all really at outdoors. But it's not simply a biological or infectious disease phenomenon. Archaeological it's it is it is it is a change in the way he was treated each other. I hope hope temporary. </s><s>[INST] I do want to say on the flip side of that. So I was mostly in boston massachusetts when the pandemic broke out. I think that's where I was. Yeah. And then I got came here to austin texas to visit my now good friend joe Rogan and he was the first person without pause. This wasn't a political statement. This was anything. Just walked toward me and give me a big hug and say, it's great to see you and I can't tell you how great it felt because I in that moment realized the absence of that connection back in boston over just a couple of over a couple of months and it's, we'll talk about it more. But it's it's tragic to think about that distancing that dissolution of common humanity at scale what what kind of impact it has on society. Just across the board. Political division. Uh, and just in the quiet of your own mind and the privacy of your own home, the depression, the sadness the loneliness that leads to suicide and forget suicide Just loki suffering. [/INST] Yeah. No, I think that's that's the suffering that isolation were not meant to live alone. Were not meant to live apart from one another. And that's of course the ideology of lockdown is to make people live apart alone isolated so that we don't spread diseases to each other. Right? But we're not actually designed as a species to live that way. And that what you're describing, I think if everyone's honest with themselves have felt, especially in places where that where lockdowns have been sort of very militantly enforced has felt deep into their core. </s>

<s>[INST] fantastically better investment. Like the thing I've been advocating, a modest proposal to eat the children of the poor in England. Was that in Jonathan Swift's modest proposal from a few centuries ago? So it's not just cost benefit, it's also in the context of what is moral and the full complexity. [/INST] You just hit on something really important. Having been on this beat for so long, and again, on the disaster beat as well, earthquakes, I can't tell you how many disaster science experts keep telling me, like everyone says, invest for preparedness. A strict cost benefit analysis will always tell you a dollar invested in resilience before a community gets hit by whatever is worth 10. You'll always have to spend 10 after. And so it's fine to do the cost benefit stuff, but it's just the baseline. Then you have to look at the social science, which shows, or history, which shows you how few times we do it. It's like we just don't do it. Therefore, you can bang that drum. Your work is valuable, but it's really constrained because show me in the world where that does happen and then how you turn that success, which is basically something not happening, into the story, into the story. So very briefly, we do this for a lot of countries. So we did it for Haiti, for instance, funded by the Canadian Development Ministry because they were basically saying, we spent a billion dollars in Haiti since the earthquake and we really can't tell the difference. So they want it to find... I mean they actually say that, right? And so they said, we want to find out what are the really smart things you can do in Haiti. And so we, together with lots of people in Haiti and all the business community and the political community and the religious community and labor community and everybody else, what are the smart things to do? And then we had economists evaluate it. There were a lot of these things that everybody wanted that were not all that smart. There's actually a lot of smart things. And yes, the politicians didn't pick most of them. So our senses, we try to give people, you're thinking about these 70 things, you should actually just think about these 20 things, and then we consider ourselves incredibly lucky if they actually do one of them. </s><s>[INST] You wrote the book How to Spend 75 Billion Dollars to Make the World a Better Place. So can we just list some of the things, if you got $75 billion, how do you spend them? [/INST] Right. So there's some incredibly good and very, very well documented things that you could spend money on. So we have two big infectious diseases that almost nobody think about, because we only think about COVID. But tuberculosis used to be the world's biggest infectious disease killer. It still kills about one and a half million people every year. The reason why we don't really worry about it is because we fixed it 100 years ago. We know how to fix it. It's just basically getting medication to people. It's also about getting them to take it while when they're sort of been cured because you need to take it for four to six months and that's actually hard to do. So you also need to incentivize that in some kind of way. It turns out it's incredibly cheap to basically save almost all of the 1.5 million people. These are people that die in the prime of their lives. They're typically parents, so it would also have a lot of knock-on effects. And basically we find for a couple billion dollars, you could save the vast number of these. Not all of them, but you could save the vast number of them. It would also improve outcomes in all kinds of other ways. </s><s>[INST] Likewise with malaria and other, it has some somewhat better PR. It's funny to think of malaria as PR and tuberculosis. They need to improve their PR department. Those mosquitoes are the good PR. [/INST] But by far the biggest infectious disease that got good PR, if you will, was HIV. And I'm not trying to compare it and say, oh, it's worse or better to have HIV than tuberculosis or anything, but I'm simply saying we are underfunding because it doesn't really get the public attention. </s><s>[INST] But spending money on that has, in terms of benefit, a much bigger impact. [/INST] So every dollar you spend on TB will probably do about $43 worth of good. So it'll do an amazing amount of good, basically because it'll save lives, it'll make sure parents stay with their kids and be more productive in their communities, and it'll have a lot of knock-on effects. And it's incredibly cheap to do. Same thing with malaria. It's mostly mosquito nets that we need to get out. </s><s>[INST] And you're saying just to contrast with climate change, the dollar you spend on, no, not climate change, but decreasing emissions does not come close to the $43 benefit? [/INST] No, nobody would ever argue that. So very, very enthusiastic climate advocates would probably say it'll do $2 or $3 worth of good for every dollar. So it's still worthwhile to do. That's what they would say. I would argue, and I think a lot of the evidence seems to side that way, that a lot of the things that we're doing deliver actually less than it all or back. But it's certainly not nearly the same kind of place. But there's many, many other things. And just if you'll allow me to- Yeah, it's crazy. I love this. Yeah. There are lots of other things, for instance, e-procurement. So it's incredibly boring. So most developing countries, well, actually most governments spend most of their money on procurement. It's typically incredibly corrupt. So we did this project for Bangladesh. Can you explain procurement? Yes. So that's governments buying stuff. So a large part of the government revenue is spent on buying anything from post-it notes to roads. And obviously roads are much, much more expensive. It's mostly infrastructure stuff. Hugely corrupt. For instance, in Bangladesh, it would already have been decided among the ruling elite in that local area who's going to get this. So they'll have this bidding competition where you have to hand in an envelope, a sealed envelope with your bid on it. But you put a goon outside the office, so you literally physically can't get in with your bid. Now, what we found, and I'm not claiming any sort of ownership to this. A lot of smart people have done this way before. We're just simply proving that it's a good idea. It turns out that if you put this on eBay, essentially, so if you do an e-procurement system where bidders can come in, suddenly it becomes harder to put up the goon. You can still do it, but it's harder to do it. It also means you get bids from all over Bangladesh. And in general, you'll get bids from all over. Actually, it turns out you get better quality. But most important is you get it much cheaper. So basically, you can simply save money. So we did a scaled experiment in Bangladesh where we had about 4% go to be e-procurement. And you could compare what it would have cost, and then what it did cost. And the average reduction was, as I remember, it's 7%. And the finance minister loved it, because that basically gives him a lot more money, or you can buy more stuff at the same cost. </s>

<s>[INST] and when you're doing that kind of back and forth on the topic like the economy, which you have to me or have some naive but it seems both incredible, An incredibly complex the economy, the trading, the transactions that these transactions between two individuals somehow add up to this giant mechanism. You've put out a 30 minute video you have a lot of incredible videos online That people should definitely watch on YouTube. But you've put out this 30 minute video titled how the economic machine works. That is probably one of the best, if not the best video I've seen on the internet in terms of educational videos, so people should definitely watch it, especially because it's not that um the individual components of the video are somehow revolutionary, but the simplicity and the clarity of the different components just makes you, there's a few light bulb moments there about what, how the economy works as a machine. So as you describe, there's three main forces that drive the economy productivity growth, short term debt cycle, long term debt cycle. The former productivity growth is how valuable things, how much value uh people create valuable things, people create. The latter is people borrowing from the their future selves to hopefully create those valuable things faster. So this is an incredible system to me, maybe we can linger on in a little bit, but you've also said, well, most people think about as money is actually credit total amount of credit in the US, $50 trillion, total amount of money is $3 trillion. That's just crazy to me. Maybe, maybe I'm sorry, maybe you can educate me, but that seems crazy. It gives me just pause that human civilization has been able to create a system that has so much credit. So that's a long way to ask, do you think credit is good or bad for society? That system of that's so fundamentally based on credit. [/INST] I think credit is great, even though people are often overdo it. Credit is that somebody has earned money and uh you know, and what happens is they lend it to somebody else who's got better ideas and they cut a deal and then that person with the better ideas is going to pay it back. And if it works well, it helps resource allocations go well, providing people like the entrepreneurs and all of those, they need capital, they don't have capital themselves, and so somebody's going to give them capital and they'll give them credit along those lines, then what happens is it's not managed well in a variety of ways. So I did another book on principles, Principles of big debt crisis is that go into that, and it's free, by the way, I put it free online um on as a pdf? So if you go online and you look principles for big debt crisis is under my name. You can download it in a pdf where you can buy a print book of it and it goes through that particular process and so you always have it overdone in always the same way everything by the way almost everything happens over and over again for the same reasons. Okay so these debt crisis has all happened over and over again for the same reasons they get it overdone. In the book it explains how you identify whether it's overdone or not. They get it overdone and then you go through the process of making the adjustments according that and then and it explains how they can use the levers and so on. If you didn't have credit then you would be sort of everybody sort of be stuck. So um credit uh is a good thing but it can easily be overdone. So now we get into the what is money, what is credit? Okay you get into money and credit. So if you're holding credit and you think that's worthwhile keep in mind that the central bank let's say it can print the money what is that problem that you have an I. O. U. And the I. O. U. Says you're going to get a certain number of dollars let's say or yen or euros and that is what the Iou is? And so the question is will you get that money and what will it be worth? And then also you have a government which is a participant in that process because they want they are on the hook, they owe money And then will they print the money to make it easy for everybody to pay. So you have to pay attention to those two. I would suggest like you you recommend other people just take that 30 minutes and it and it and it comes across pretty clearly. But my conclusion is that of course you want it. And even if you understand um it and the cycle as well you can benefit from those cycles rather than to be hurt by those cycles because I don't know the way the cycle works is somebody gets over indebted they have to sell an asset. Okay then I don't know that's when assets become cheaper. How do you acquire the asset? It's a whole process. </s><s>[INST] So again maybe another dumb question. But uh [/INST] there are no such things as dumb questions. There you go. But what is money? </s>

<s>[INST] Can you look on that? What do you mean? I missed where you're placing consciousness in the second? [/INST] Well, so, so if you are a computational is do you think that what creates consciousness is the implementation of a computation? </s><s>[INST] Some property emergent property is a computation itself. [/INST] Yeah, yeah, you could say that. But then the question is which what what's the class of computations such that when they are run consciousness emerges. So if you just have like something that adds one plus one plus one plus one, like a simple computation, you think maybe that's not going to have any consciousness if on the other hand, the competition is one like our human brains are performing. Where as part of the computation there is like, you know, a global workspace, a sophisticated attention mechanism. There is like self representations of other cognitive processes and a whole lot of other things that possibly would be conscious. And in fact, if it's exactly like ours, I think definitely it would, but exactly how much less than the full competition that the human brain is performing would be required? Uh is a little bit, I think often open question. Um and he asked another uh interesting question as well, which is would it be sufficient to just have, say the brain or would you need the environment? </s><s>[INST] Right, That's a nice [/INST] in order to generate the same kind of experiences that we have. And there is a bunch of stuff we don't know. I mean, if you look at, say current virtual reality environments, one thing that's clear is that we don't have to simulate all details of them all the time in order for, say the human player to have the perception that there is a full reality that you can have, say procedurally generated, which we might only render a scene when it's actually within the view of the player character. Um And so similarly, if this if this if this environment that we perceive is simulated, it might be that all of the parts that come into our view are rendered at any given time. And a lot of aspects that never come into view, say the details of this microphone, I'm talking into exactly what each atom is doing at any given point in time might not be part of the simulation, only a more coarse grained representation. </s><s>[INST] So that that, to me is actually from an engineering perspective, why the simulation hypothesis is really interesting to think about is how much how difficult is it to fake sort of in a virtual reality context? Don't know fake is the right word, but to construct the reality that it's sufficiently real to us to be uh to be immersive in a way that the physical world is I think that's just that's actually probably an answerable question of psychology, of computer science, of how, how where is the line where it becomes so immersive that you don't want to leave that world? [/INST] Yeah. That you don't realize while you're in it that it is a virtual world. </s><s>[INST] Yeah, those are two actually questions yours is the more sort of the good question about the realism but mine from my perspective what's interesting is it doesn't have to be real but um how how can you construct the world that we wouldn't want to leave? [/INST] Uh Yeah, I mean I think that might be too low a bar. I mean if you think say when when people first had pong or something like that, I'm sure there were people who wanted to keep playing it for a long time because it was fun and I wanted to be in this little world. Yeah, I'm not sure what I would say. It's immersive. I mean I guess in some sense it is but like an absorbing activity, it doesn't even have to be. </s><s>[INST] But they left that world though. That's so like I think that bar is um deceivingly high so they eventually so they you can play pong or Starcraft or would have more sophisticated games for hours for four months, you know while. Well the work has to be in a big addiction but eventually they escape [/INST] that. You mean when it's uh absorbing enough that you would spend your entire, it would choose to spend your entire life in there and </s><s>[INST] then thereby changing the concept of what reality is your reality. Your reality becomes the game, not because you're fooled, but because you've made that choice. [/INST] Yeah. It may be different. People might have different preferences regarding that. Some some might uh even if you had any perfect virtual reality might still prefer not to spend the rest of their lives there. I mean in philosophy there's this experience machine thought experiment. Have you come across this? So robert Nozick had this thought experiment where you imagine some crazy uh super duper neuroscientists of the future have created a machine that could give you any experience you want if you step in there. Um and for the rest of your life, you can kind of pre programmed it in different ways. Um So you're, you're fund the dreams could come true, you could, whatever you dream, you want to be a great artist, a great lover, like have a wonderful life, all of these things, if you step into the experience machine will be your experiences constantly happy. And um but would you kind of disconnect from the rest of your life and you would float there in the tank. Um and the Nozick thought that most people would choose not to enter the experience machine. Many might want to go there for a holiday, but I wouldn't want to have to check out of existence permanently. And so he thought that was an argument against um certain views of value according to what we what we value is a function of what we experience. Because in the experience machine you can have any experience you want. And yet many people would think that would not be much value. So therefore what we value depends on other things than what we um experience. So, </s>

<s>[INST] 1000 by 1000 by 1000 [/INST] Adams </s><s>[INST] that's a really clean way of putting [/INST] It. Well fan like a modern transition. If you look at the fan it's like 120 atoms wide. But we can make that thinner and then there's there's a gate wrapped around it and then their spacing, there's a whole bunch of geometry and you know, a competent transistor designer could count both Adams in every single direction. Uh huh. Like there's techniques now to already put down atoms in a single atomic layer. Right? And you can place Adams if you want to it's just you know, from a manufacturing process placing Adam takes 10 minutes and you need to put you know 10 to the 23rd atoms together to make a computer. It would take a long time. So the the methods are you know, both shrinking things and then coming up with effective ways to control what's happening, </s><s>[INST] manufacture stable and cheaply. [/INST] Yeah. So so the innovation stocks pretty broad, you know there's there's equipment, there's optics, there's chemistry, there's physics, there's material science, there's metallurgy, there's lots of ideas about when you put them from materials together, how they interact with a stable, stable over temperature, you know like are the repeatable, you know, there's like there's like literally thousands of technology is involved. </s><s>[INST] But just for the shrinking, you don't think we're quite yet close to the fundamental limits of physics. [/INST] I did a talk on moore's law and I asked for a road map to a path about 100 and after two weeks They said we only got 2, 50, </s><s>[INST] 100, extra 100 x drink [/INST] 50. And I said, Why don't you give it another two weeks? Well, here, here's the thing about moore's law, right? So I believe That the next 10 or 20 years of shrinking is going to happen right now as a computer designer, you have two stances. You think it's going to shrink, in which case you're designing and thinking about architecture in a way that you'll use more transistors or conversely not be swamped by the complexity of all the transistors you get right, you have to have a strategy, </s><s>[INST] you know, so you're open to the possibility and waiting for the possibility of a whole new army of transistors ready to work. [/INST] I'm expecting expecting more transistors every two or three years by a number large enough that how you think about design, how you think about architecture has to change. Like imagine you're, you you build build brick buildings out of bricks and every year the bricks are half the size or every two years. Well, if you kept building bricks the same way, you know, so many bricks per person per day, the amount of time to build a building when we go up exponentially. Right? But if you said I know that's coming. So now I'm going to design equipment and moves bricks faster, uses them better because maybe you're getting something out of the smaller bricks more strength in their walls, you know, less material efficiency out of that. So once you have a road map with what's going to happen, transistors they're gonna get we're gonna get more of them. Then you design all this collateral around it to take advantage of it and also to cope with it. Like that's the thing people understand, it's like if I didn't believe in moore's law and moore's law, transistors showed up, my design teams are all drowned. </s><s>[INST] So what's the what's the hardest part of this in flood of new transistors? I mean even if you just look historically throughout your career, what's what's the thing what fundamentally changes when you had more transistors in the task of designing and architecture? [/INST] Well there's there's two constants, right? One of those people don't get smarter, </s><s>[INST] by the way, there's some signs showing that we do get smarter but because the nutrition whatever. [/INST] Sorry to bring that in effect. Yeah, familiar with it, nobody understands it, nobody knows if it's still going on, so that's on it </s><s>[INST] or whether it's real or not. But yeah, it's a [/INST] I sort of anyway, but I believe for the most part of people aren't getting much smarter, </s><s>[INST] the evidence doesn't support it, right [/INST] and then teams can't grow that much. Alright, alright, so human beings Under, you know, we're really good in teams of 10, you know, up to teams of 100 they can know each other beyond that. You have to have organizational boundaries. So you're kind of you have those are pretty hard constraints right? So then you have to divide and conquer like as the designs get bigger, you have to divide it into pieces, you know, the power of abstraction layers is really high. We used to build computers out of transistors. Now we have a team that turns transistors and logic cells and another team that turns them into functional. It's another one that turns into computers. All right. So we have abstraction layers in there and you have to think about when you shift gears on that. We also use faster computers to build faster computers. So some algorithms run twice as fast on new computers but a lot of algorithms are in squared. So a computer with twice as many transistors and it might take four times times as long to run. So you have to re factor the software like simply using faster computers to build bigger computers doesn't work. So you have to think about all these things. </s>

<s>[INST] just like you said, it's also on us. I try to be all about love and compassion, empathy on twitter. I mean one of the things not picking is a fascinating term, one of the things that people do that's I think even more dangerous is not picking applied to individual statements of good people. So basically worst case analysis in computer sciences taking sometimes out of context, but sometimes in context. Uh a statement, one statement by a person like I've been, because I've been reading the rise and fall of the third Reich I've often talked about Hitler on this podcast with folks and it is so [/INST] easy, but </s><s>[INST] I'm all leaning in, I'm 100% because well it's actually a safer place than people realize because it's history and history in long form is actually very fascinating uh to think about and it's but I could see how that could be taken totally out of context. It's very [/INST] worrying about, you know, these digital infrastructure, not just to send me things, but they're sort of permits or anything you say at some point, someone can go back and find something you said three years ago, perhaps jokingly, perhaps not, maybe you're just wrong and you know, and like that becomes they can use that to define you if they have content and we all need to be able more forgiving. I mean somewhere in my twenties, I I told you I was going through all my different friends and I was like, you know, every one of them has at least like one nutty opinion. Yeah, like there's like nobody who's like completely except me of course, but I'm sure they thought that about me too and uh and so you just kind of like learn to be a little bit tolerant that like okay, there's just uh </s><s>[INST] yeah, I wonder who the responsibility lays on there. Like I think ultimately it's about leadership like the previous President Barack Obama has been I think quite eloquent at walking this very difficult line of talking about cancel culture, but it's difficult, it takes skill because you say the wrong thing and you piss off a lot of people and so you have to do it well. But then also the platform of the technology is should slow down create friction and spreading this kind of not picking in all its forms [/INST] absolutely. You know, and your point that we have to like learn over time how to manage it and we can put it all on the platform and say you guys design it because if we're idiots about using it, you know, nobody can present a platform that with stands at and and every new technology, people learn it's dangerous, you know, when someone invented fire great cooking and everything, but then somebody burned himself and then you had to like learn how to like avoid maybe somebody invented a fire extinguisher later and what's so you kind of like figure out ways of working around these technologies, someone invented seatbelts etcetera. Um And that's certainly true with all the new digital technologies that we have to figure out not just um technologies that protect us, but ways of using them that um emphasize that are more likely to be successful than dangerous. </s><s>[INST] So you've written quite a bit about how artificial intelligence might change our world. How do you think if we look forward again, it's impossible to predict the future. But If we look at trends from the past and we tried to predict what's going to happen in the rest of the 21st century, how do you think Ai will change our world? [/INST] A big question. You know, I'm mostly a techno optimist. I'm not at the extreme, you know the singularity is near the end of the spectrum, but I do think that we are likely in for some significantly improved living standards, some really important progress. Even just the technologies that are already kind of like in the can that haven't defused, You know when I talked earlier about the J curb it could take 10 2030 years for an existing technology to have the kind of profound effects. And when I look at whether it's vision systems, voice recognition, problem solving systems. Even if nothing new got invented, we would have a few decades of progress. So I'm excited about that and I think that's gonna lead to us being wealthier, healthier. I mean the healthcare is probably one of the applications that I'm most excited about. Um so that's good news, I don't think we're gonna have the end of work anytime soon. Um There's just too many things that machines still can't do when I look around the world and think of whether it's it's child care or healthcare, clean the environment, um interacting with people, scientific work, artistic creativity. These are things that for now machines aren't able to do nearly as well as humans, even just something as mundane as you know, folding, laundry or whatever. And many of these I think are going to be years or decades before machines catch up. You know, I may be surprised on some of them, but overall I think there's plenty of work for humans to do. There's plenty of problems in society that need the human touch, so we'll have to repurpose we'll have to As machines are able to do some tasks, people are going to have to re scale and move into other areas. Um and that's probably what's gonna be going on for the next, you know, 10 2030 years or more kind of big restructuring of society will get wealthier and people will have to do new skills Now, if you turn that down further, I don't know 50 or 100 years into the future then you know, maybe all bets are off then it's possible that that machines will be able to do most of what people do. You know, say one or 200 years. I think it's even likely. And at that point then we're more in the sort of abundance economy. Then we're in a world where there's really little for the humans can do economically better than machines other than be human and you know, well that will take a transition as well. Kind of more of a transition of how we get meaning in life and what our values are. But shame on us if we screw that up. I mean that should be like great, great news and it kind of saddens me that some people see that as like a big problem. You know, I think I would be should be wonderful if people have all the health and material things that they need and can focus on loving each other and discussing philosophy and playing and doing all the other things that don't require work. </s>

<s>[INST] Don't tell me you're gonna turn this [/INST] into a machine learning problem. So then you turn into a machine learning problem, And then you have a space of genetic algorithms and reinforcement learning and all these cool things. </s><s>[INST] Can you include that into the stack, into the modular stack? Yeah, yeah. Where does it sit? Where does it live? [/INST] Is it a separate thing or is it part of the compilation? Yeah, yeah. So you start from simple and predictable models. And so you can have full control and you can have coarse-grained knobs like nudged systems, so you don't have to do this. But if you really care about getting the best, last ounce out of a problem then you can use additional tools and they're the cool things you don't want to do this every time you run a model you want to figure out the right answer and then cash it and once you do that you can get you can say okay cool I can get up and running very quickly I can get good execution out of my system I can decide if something's important and if it's important I can go throw a bunch of machines at it and do a big expensive search ever the space using whatever technique I feel like its pretty up to to the problem, and then when I get the right answer, cool, I can just start using it. And so you can get out of this trade-off between, okay, am I gonna spend forever doing a thing, or do I get up and running quickly, and as a quality result, these are actually not in contention with each other if the system's designed to scale. </s><s>[INST] You started and did a little bit of a whirlwind overview of how you get 35,000X speed up or more over Python. Jeremy Howard did a really great presentation about sort of the basic, like, look at the code, here's how you get the speed up. Like you said, that's something we could, probably developers can do for their own code to see how you can get this gigantic speed up. But can you maybe speak to the machine learning task in general, how do you make some of this code fast and specific, like, what would you say is the main bottleneck for machine learning tasks? So are we talking about that mall matrix multiplication, how do you make that fast? [/INST] So, I mean, if you just look the Python problem, right, you can say, how do I make Python faster? There's been a lot of people that have been working on the, okay, how do I make Python 2X faster, 10X faster, or something like that, right? And there've been a ton of projects in that vein, right? Mojo started from the, what can the hardware do, like, what is the limit of physics? What is the speed of light, what is it? Like, how fast can this thing go? And then, how do I express that? Yeah, right. And so it wasn't well, it wasn't anchored relatively on make Python a little bit faster. It's saying cool I know what the hardware can do. Let's unlock that right now when you yeah </s><s>[INST] Yeah, just say how how gutsy that is to be in the meeting and as opposed to trying to see how do we get the improvement? [/INST] It's like what can the physics do? I Mean, maybe I'm a special kind of nerd, but you look at that. What is the limit of physics? How fast can these things go, right? When you start looking at that, typically it ends up being a memory problem. Mm-hmm. Right and so today Particularly with these specialized accelerators. The problem is that you can do a lot of math within them But yet you get bottleneck sending data back and forth to memory whether it be local memory or distant memory or disk or whatever It is and and that that bottleneck particularly is the training sizes get large as you start doing tons of inferences all over the place Like that becomes a huge bottleneck for people, right? So again, what happened is we went through a phase of many years where people took the special case and hand tuned it Tweaked it and tricked out and they knew exactly how the hardware worked and they knew the model and they made it they made it fast Didn't generalize and so you can make you know Resident 50 or some or Alex net or something inception v1 like you can you can do that, right because the models are small They fit in your head Right, but as the models get bigger more complicated as the machines get more complicated it stops working Right, and so this is where things like kernel fusion come in. So what is kernel fusion? This is this idea of saying let's avoid going to memory and let's do that by building a new hybrid Kernel and a numerical algorithm that actually keeps things in the accelerator instead of having to write all the way out to memory All right, what's happened with with these accelerators now is you get multiple levels of memory like in a GPU For example, you'll have global memory and local memory and like all these things If you zoom way into how hardware works the register file is actually a memory so the registers are like an L0 cache and so a lot of Taking advantage of the hardware ends up being full utilizing the full power in all of its capability and this has a number of problems, right one of which is again the complexity of disaster, right? There's too much hardware. Even if you just say let's look at The chips from one line of vendor like Apple or Intel or whatever it is Each version of the chip comes out with new features and they change things so that it takes more time or less time to do different Things and you can't rewrite all the software Whenever a new chip comes out, right? And so this is where you need a much more scalable approach and this is what mojo and what the Modular stack provides is it provides this infrastructure and the system for factoring all this complexity and then allowing people to express algorithms You talk about auto tuning for example Express algorithms in a more portable way so that when a new chip comes out you don't have to rewrite it all So to me like, you know, I kind of joke like what is a compiler? Well, there's many ways to explain that you convert thing a and thing B and you convert source code to machine code Like you can talk about many many Things that compilers do but to me it's about a bag of tricks It's about a system and a framework that you can hang Complexity it's a system that can then generalize and it can work on problems that are bigger than fit in one human's head Right. And so what that means what a good stack and what the modular stack provides is The ability to walk up to it with a new problem and it'll generally work quite well And that's something with a lot of machine learning infrastructure and tools and technologies don't have Typical state-of-the-art today as you walk up particularly if you're deploying if you walk up with a new model You try to push it through the converter and the converter crashes. That's crazy The state of ML tooling today it's not anything that a C programmer would ever accept, right? And it's always been this kind of flaky side of tooling that's never been integrated. Well, and it's been Never worked together and because it's not designed together. It's built by different teams It's built by different hardware vendors is built by different systems it's built by different internet companies They're trying to solve their their problems Right and so that means that we get this fragmented terrible mess of complexity </s>

<s>[INST] and [/INST] now they have some value, which could be glory. I mean, this is the story of the kings and to some lesser extent, the queens of europe for hundreds of years was it was basically a contest and it was the more was the sport of kings and to some extent, they were just seeking status through violent competition and they paid a lot a big price out of the royal purse. But but uh but they didn't care most of the suffering. Um and and so they were too quick to go to war. And so that's I think that detachment of leaders combined with, you know, you mingle it with this, that one bucket that uncheck Nous and you mingle that with the fact that leaders might have one of these values noble or otherwise that carry them to war combined to explain a good number of conflicts as well. And that's a good illustration of why I think like autocracy and unaccountable power is I could make that story for all of those things, all five buckets, they're all we're all more susceptible to these things. All five of these things when leaders are not accountable to the people and their group and and that's what makes it like the meta for me the meta cause of of conflict in all of human history and sadly today </s><s>[INST] does the will to power play into this. The desire for power like that's a human thing. Again, in the calculation that shall we put that in the misperceptions bucket or is it as misperceptions essentially about interaction between humans and powers more about the thing you feel in your heart when you're alone as a, as a leader. [/INST] You know, I said there were three strategic reasons like the Czech leaders, the commitment problems, uncertainty there to sort of more psychological and I called them intangible incentives and misperceptions. The way that like a game theorist or the way that behavioral economists think about those two was just to say preferences and then erroneous beliefs and mistakes is like so the our preferences are preferences, right? And so utility functions, whatever we want to call it. Like there's no that's why we wouldn't call them a misperception or rationality. We want. We like what we like if we like power, if we like relative status, if we like if we like our racial purity, if we like our liberty, if we like whatever it is that we have convinced ourselves, we value </s><s>[INST] maybe you fell in love with the rival queen king. [/INST] Exactly when I said it was a big bucket full of stuff that rhymes like that, That's a pretty messy bucket. Like there's a lot of different stuff in there. And I'm just trying to say like let's be clear that just about the lodge the shared logic of these things is maybe just, you know, they're really dissimilar, but let's be clear about the shared logic. Uh and if it were true that deep down we were aggressive people who just liked violence and enjoyed the blood </s><s>[INST] or [/INST] some percentage of us do that would be there too. Um and so I just want to say that's but you know, we're really quick to recognize those right? When we diagnose the war as an armchair analyst or as a journalist or something. We really jumped to those. We don't need a lot of help to like see those happening. So we probably put a little bit too much emphasis on them is maybe the only thing that I would caution because we're the others are more subtle and they're often there and they contribute. </s><s>[INST] So I just to look at something he said before, would it be accurate to say when the leaders become detached from the opinion of the people is that's more likely to lead to war? So, uh, [/INST] and mechanically, it's just they're, they're going to bear fewer costs. So it's gonna, it's going to basically narrow the set of deals that they're going to be willing to accept instead of violence at the same time. Most of the time, </s><s>[INST] it's [/INST] not enough because the leaders still bear a lot of costs of war. You could be deposed, you could be killed, you can be tried and the public purse is going to be empty. That's like the one story throughout history is at the end of the day you're regime is broke as a result of war. And so you still internalize that a little bit. Um, if I had to say like, you know, in my three buckets or through my bucket so far, I've sort of started with like Ukrainian and transitions. And then I jumped and then I said the essentially then you really have to understand Russian autocracy for them just to understand why they would ask something so cruel. But I mean I think the uncertainty is really important here as well. Like if you think of it like think of all the things the way this has played out and just in some ways how many, how many, how many ways we have been surprised? We've been surprised by the unity and the coherence of the West and the sanctions, that sort of what's happened is it was in the realm of possibility, but it was sort of like the best case scenario from the perspective of the West and the worst case scenario for the Russians. The second thing is just the pluckiness and the effectiveness and the intransigence and the nobility of this Ukrainian resistance that's again was within the realm of possibility, but wasn't necessarily the likely thing right? Was again maybe the worst realization for Russia. The best realization in some sense for in terms of revealed strength and resolve. Um and then the other thing that's been revealed is just how like the from the corruption and ineptitude and problems on the Russian military side, them again within the realm of possibility. Maybe people who really knew the Russian military are less surprised than the rest of us but but also one of the worst possible draws to Russia and so Putin asking this terrible price and expecting Ukraine to roll over or the Western roll over at least to a degree was based on like a different set of probably was based on just expecting something in the middle of the probability distribution and not one of all these different tail events. And so the fact that the world is so uncertain and the fact that Putin can come with a different set of expectations than the Ukrainians in the West and all these players can just have a hard time agreeing on just what the facts are. Because we live in an uncertain world. Everyone is quick to say, oh we miscalculated. Well, I'm not, I don't know if he miscalculated. I think he just he got a really bad draw on in terms of the what the realized outcomes are here. And so that I mean good for everybody else in some sense except you know, the fact that it's involving a lot of violence is the tragedy so </s>

<s>[INST] so so the neural net is the architecture for the background model, the base thing and then everything else is just on top of that. [/INST] It's not just the text, right? It's you're giving it gadgets and gizmos. So uh you're teaching some kind of special language by which we can, we can save arbitrary information and retrieve it at a later time and you're, you're telling about the special tokens and how to arrange them to use these interfaces. It's like, hey you can use a calculator, here's how you use it. Just do 53 plus 41 equals and when equals is there a calculator will actually read out the answer and you don't have to calculate it yourself and you just like tell it in english this might actually work, </s><s>[INST] do you think in that sense Gado is interesting. The deepmind system that it's not just the language but actually throws it all uh in the same pile, images, actions, all that kind of stuff that's basically what we're moving towards. Yeah, [/INST] I think so. So Gatto is, is very much a kitchen sink approach to like um reinforcement learning lots of different environments with a single fixed transformer model. Right. Um I think it's a very sort of early result in that in that realm. But I think yeah, it's along the lines of what I think things will eventually look like. </s><s>[INST] Right? So this is the early days of a system that eventually will look like this. Like from a rich, rich, sudden perspective. [/INST] I'm not super huge fan of, I think all these interfaces that look very different. Um I would want everything to be normalized into the same ap I so for example, screen pixels very same api instead of having like different world environments that are very different physics and joint configurations and appearances and whatever and you're having some kind of special tokens for different games that you can plug. I'd rather just normalize everything to a single interface. So it looks the same to the neural net, if that makes sense. So it's </s><s>[INST] all gonna be pixel based pong in the end. [/INST] I think so, okay, </s><s>[INST] uh let me ask you about your own personal life. A lot of people want to know you're one of the most productive and brilliant people in the history of ai what is a productive day in the life of Andre Karpati look like. What time do you wake up? Because imagine um some kind of dance between the average productive day and a perfect productive day. So the perfect productive day is the thing we strive towards and the average is kind of what it kind of converges to giving all the mistakes and human eventualities and so on. So what time do you wake up? Like a morning person? [/INST] I'm not a morning person, I'm a night owl for sure </s><s>[INST] it's stable or not, [/INST] it's semi stable like eight or nine or something like that. During my PhD it was even later I used to go to sleep usually at three am I think uh am hours are are precious and very interesting time to work because everyone is asleep um at at eight AM or seven AM. The East coast is awake. So there's already activity, there's already some text messages, whatever there's stuff happening, you can go on like some news website and there's stuff happening distracting at three a.m. Everything is totally quiet and so you're not gonna be bothered and you have solid chunks of time to do work. Um So I like those periods night all by default and then I think like productive time basically um what I like to do is you need you need to like build some momentum on the problem without too much distraction and um you need to load your RAM uh your working memory with that problem and then you need to be obsessed with it when you're taking a shower, when you're falling asleep. You need to be obsessed with the problem and it's fully in your memory and you're ready to wake up and work on it right there. So </s><s>[INST] it is the scale of uh is this in a scale temporal scale of a single day or a couple of days a week? A month. [/INST] So I can't talk about one day basically in isolation because it's a whole process when I want to get when I want to get productive in the problem. I feel like I need a span of a few days where I can really get in on that problem and I don't wanna be interrupted and I'm going to just be completely obsessed with that problem and that's where I do most of my good workouts. </s><s>[INST] You've done a bunch of cool like little projects in a very short amount of time very quickly so that that requires you just focusing on it. [/INST] Yeah basically I need to load my working memory with the problem and I need to be productive because there's always like a huge fixed cost to approach any problem. Uh you know like I was struggling with this for example at Tesla because I want to work on the small side project but okay you first need to figure out okay I need to associate into my cluster, I need to bring up a V. S. Code editor so I can work on this. I need to I run into some stupid error because of some reason like you're not at a point where you can be just productive right away you are facing barriers and so it's about really removing all that barrier and you're able to go into the problem and you have the full problem loaded in your memory </s>

<s>[INST] It's very certain just you said that there's a bunch of them that have done because there's some conspiracy theories that are dramatic and true. The question is, I wish there was a way to prove that some of them are not. And it's very difficult because so much a shroud of mystery. Like one, you know, one of the things that bothered by is when people accuse other like athletes of using steroids for example. And it's just, yes, a lot of people use steroids, but it sucks the, that people just don't believe you like uh there's some incredible athletes that look shredded. They look just incredible performers and everybody just says that they're on steroids that kind of assumed. I [/INST] mean that people accuse me all the time of being on performance enhancing drugs and steroids. And it is hard. But what I remind them is it's what my my my my my my appearance as a result of dedication. But the hard work, diet, exercise, </s><s>[INST] dedication, keto. [/INST] Uh, I'm on, I'm on key. I'm doing a version of your keto. Right? So I'm doing a version of keto right now with bread and it's, you see what I mean? Your car bulb. It's, you know, with sugar. It's called keto plus sugar. And it's a very, it's a good diet for I grew up in the nineties when nobody ever lost weight sadly because every diet was like, you can eat what you want, just be accountable. No one even knew what that man. So it would be like my mother being like, if you have chocolate chip pancakes, have a glass of </s><s>[INST] water, [/INST] just take a walk around the block. You're going to Mcdonalds three times a day, just walk around the block. It's my parents, my mother just walk around the block, you're fine. Um, I'm going to have a cigarette walk, 20 steps, walk 20 steps back, it's exercise. So, um, you know, uh, there's too many conspiracies after a lot of them are true. A lot of them are bitter angry people trying to justify their own, uh, impotence and not being able to do anything in life. And they're like, the people that have done something in life, they're all nefarious. It's all the car just act against me. That's 100% true. 100%. It attracts usually people that have not figured out a way to succeed and, or I haven't succeeded on the level that they want to, but that also being true. There are, there is a fair amount of factory going on and provable and you know, we just have to, I think separate, uh, separate know that these things are often inflated or not true, but know that sometimes they are true. Otherwise they wouldn't exist if there was no, if there was nothing to JFK, there was nothing to 9 11 if there was, if people felt like they were being dealt with, honestly, this wouldn't exists. This exists because there are real questions that people have that don't get answered for whatever reason. And then the vacuum of the refusal to answer those questions that that information vacuum is filled with people like Alex jones who are curious and sometimes they're right and sometimes they're horribly wrong and sometimes they're all over the place and </s><s>[INST] good storytellers and people love stories and when there's an absence of actual Alex is [/INST] a uniquely american person, like a very interesting, I don't know how many countries, like how many people make a living as a conspiracy theory is a good living in other countries, right? It's very rare, right? I mean, it's very interesting and he became like, I know people that knew him, he was a kid because I go to Austin and perform a lot and you know, he was a guy that would take a bullhorn and yellow cops because he thought d we checkpoints were unconstitutional. That's what he was doing in college and he just went through, he was hated by the right, He was hated by the bush people. He was hated by the and he went from being this this this guy that was considered like a leftist, even like, even though he's never left us, he was considered this like the enemy of mainstream conservatism, like he was not and you consider the guy that wasn't a patriot wasn't this wasn't that and he just wow, like he winds and winds and ends up just being this confidant of a republican president, very divisive Republican president and he becomes as populist and everything like that. It's really wild to watch that. But I mean I do think he should retire eventually just so we could get you know some, I don't know, it seems like a it's a lot to keep doing. </s><s>[INST] Well. I hope this world allows for Alex jones to continue having a voice because just like you said, he's uh use the word fun but really he shakes up the norms of our discourse. [/INST] I do too. I do think we need to put more value. I think entertainment. You know, we knew we do need to say that like there are people that should be allowed to have a voice for entertainment purposes, right? And that's part and that's part of what donald trump now that he's not the president, come on, let the guy let him talk. </s><s>[INST] Who do you think is the best comedian of all time? [/INST] Oh that's a great question. Greatest of all time. </s><s>[INST] You mentioned Carlin, your uncle's like and Carlin. [/INST] Well Carlin is great, Carlin is really hard to argue with, but chappelle is also really great um louis C. K. Is really great. I don't know that there's what joan Rivers, it's great. I don't know, you smile at that a beast of a </s><s>[INST] comic. I'm not aware of her stand up actually assist of a comic [/INST] asked Rogan, ask any of them, um, Kennison is great. </s>

<s>[INST] them, [/INST] I have to allow myself to change and be changed by them in the case of Hunger </s><s>[INST] Ward. You [/INST] may remember </s><s>[INST] the girl um Emma [/INST] um who's </s><s>[INST] The the 10 year old girl who we [/INST] showcase in </s><s>[INST] in Auden [/INST] in the south of the country. And um </s><s>[INST] you know we [/INST] we were there when she </s><s>[INST] was admitted to the [/INST] hospital </s><s>[INST] and When she was admitted, you know this 10 year old [/INST] Girl weighed 24 lb </s><s>[INST] and she could barely stand up. [/INST] And um we </s><s>[INST] started, you know, [/INST] with the permission of the </s><s>[INST] family to start to document [/INST] um her </s><s>[INST] treatment [/INST] and to see what would happen with this young girl who was so severely malnourished and we watched her be treated by the nurses and the doctors </s><s>[INST] um in Sadak a [/INST] hospital. And </s><s>[INST] slowly over [/INST] the course of a couple of </s><s>[INST] weeks we [/INST] saw her change. </s><s>[INST] We saw [/INST] her start to sort of </s><s>[INST] gain [/INST] strength and start to recover. And she also watched the </s><s>[INST] caregivers [/INST] very carefully and I watched her watch them. And </s><s>[INST] um [/INST] I'll never forget, there was a moment where </s><s>[INST] um [/INST] about 2.5 weeks I think </s><s>[INST] into her treatment, we walked into a room [/INST] and I saw </s><s>[INST] her [/INST] offering a cap full of </s><s>[INST] water [/INST] to another younger child who was also starving, right? The shots actually in the film. </s><s>[INST] And and so to see [/INST] um Emma, this child who is starving giving sustenance to a younger, more </s><s>[INST] vulnerable child who is [/INST] also starving moved </s><s>[INST] me deeply, [/INST] right? So, I saw her learn from the caregivers around her, right? And as a human being, as a filmmaker, I was </s><s>[INST] incredibly [/INST] inspired by OEM </s><s>[INST] Emma, that capacity for compassion [/INST] is there </s><s>[INST] Even within a 10 year old [/INST] girl who's starving. </s><s>[INST] Right? And [/INST] so so you </s><s>[INST] asked what changed me? [/INST] Um that's one moment, right? I I </s><s>[INST] rather than being [/INST] crushed by such heavy content. It was actually </s><s>[INST] the opposite where [/INST] I came away inspired by a 10 year </s><s>[INST] old girl [/INST] and you know, I didn't anticipate </s><s>[INST] that, I didn't [/INST] think that's what this content would do, but it's what it did. It it </s><s>[INST] reinforced for [/INST] me, sort of this incredible capacity we all have as human </s><s>[INST] beings right [/INST] to do good, right to even within the most difficult circumstances to choose </s><s>[INST] who we become [/INST] and what we do </s><s>[INST] and and [/INST] And a 10 year old girl taught me </s><s>[INST] that will reinforce that for [/INST] me. Were you able </s><s>[INST] to [/INST] feel </s><s>[INST] the culture of [/INST] the people so </s><s>[INST] that the language [/INST] barrier, </s><s>[INST] were you able to break [/INST] through the language barrier? </s><s>[INST] The culture barrier [/INST] to </s><s>[INST] understand the people, [/INST] you know? Um </s><s>[INST] because even [/INST] even suffering has a language of of sorts, depending on where </s><s>[INST] you are, the way people [/INST] joke about things, </s><s>[INST] the way they cry, the [/INST] way this is an interesting </s><s>[INST] thing I actually want to ask [/INST] you, sorry, I'm asking </s><s>[INST] a million questions. I find [/INST] that the people, you know, I've been talking to people in Ukraine and </s><s>[INST] Russia [/INST] in general, </s><s>[INST] I've gotten a chance to [/INST] talk to people who've </s><s>[INST] been through [/INST] trauma in their life and there's a humourous </s><s>[INST] they [/INST] have about trauma and </s><s>[INST] hard times. [/INST] Um it depends on the </s><s>[INST] culture, of course, [/INST] certainly </s><s>[INST] Russian speaking [/INST] folk, I </s><s>[INST] mean the [/INST] more suffering you've experienced for some reason, the more they joke about it, it's almost like </s><s>[INST] they're able to see something deep about humanity [/INST] now that they have suffered. And they're able to laugh at the absurdity </s><s>[INST] of the injustice of it all. And you know, you [/INST] could also say it's a way for them to deal </s><s>[INST] with it but [/INST] that that humor has a </s><s>[INST] kind of profound [/INST] like um understanding </s><s>[INST] within it [/INST] about what it means to be human, </s><s>[INST] that I just and and then you to really understand [/INST] that you have to know the </s><s>[INST] language. So [/INST] I guess </s><s>[INST] I'm asking [/INST] were you able to </s><s>[INST] really feel [/INST] the humans </s><s>[INST] on the other side of the language? I'd [/INST] like to think so. I mean I mean as you </s><s>[INST] noted, you know, there [/INST] there are universals in life that that transcend language, </s><s>[INST] right? I mean [/INST] suffering is suffering, Love is love, compassion. It doesn't take place only through language. </s><s>[INST] Right? [/INST] Um it's through actions and so um was there a language bearer? </s><s>[INST] Absolutely right. [/INST] Did we try to bridge that through </s><s>[INST] through other means [/INST] and and and sort of </s><s>[INST] universal [/INST] emotions and experiences? Absolutely. That's </s><s>[INST] one of the things I always think [/INST] about when I'm filming is is how </s><s>[INST] do we distill down [/INST] to Universal's right? Um through through imagery Right? Through </s><s>[INST] um through the [/INST] vocabulary of </s><s>[INST] cinema. Right? Because I believe so deeply [/INST] that that vocabulary should be visual. </s><s>[INST] Right? So the [/INST] words, what's </s><s>[INST] the most powerful way [/INST] to express </s><s>[INST] the universal, is it [/INST] visual? Or is it a </s><s>[INST] language words, [/INST] I think it's visual and we're talking about the human </s><s>[INST] face or [/INST] human face, human </s><s>[INST] body, everything through [/INST] actions as well, </s><s>[INST] actions. The [/INST] dynamic, I'm thinking about a woman named Salha </s><s>[INST] in the film [/INST] who isn't named </s><s>[INST] but [/INST] she's you see her multiple </s><s>[INST] times throughout [/INST] the film and she's basically the </s><s>[INST] matron of the ward in this house [/INST] and she she's </s><s>[INST] the gatekeeper for the ward. [/INST] So no one enters that word </s><s>[INST] without her. She's literally the gatekeeper at the door. So [/INST] No one comes in unless Salha allows them to come in. </s><s>[INST] Right? But then she [/INST] also is sort of </s><s>[INST] like [/INST] the first point of contact for compassion in the </s><s>[INST] ward. So [/INST] when when mothers and families are </s><s>[INST] admitted, she [/INST] forms relationships between the moms and the grandmothers, for example, who are admitted in her living there on the </s><s>[INST] ward and [/INST] she does it through hugging, </s><s>[INST] right? She [/INST] does it </s><s>[INST] through bringing them [/INST] food </s><s>[INST] and she forms [/INST] these </s><s>[INST] really rather quickly deep [/INST] relationships of compassion with the </s><s>[INST] families. And [/INST] so it's amazing to watch and no languages needed, right to bear witness to this. </s><s>[INST] And [/INST] and she also </s><s>[INST] suffers because of that. Right? [/INST] And so at the near the end of the film, if you </s>

<s>[INST] That's a tricky question because a lot of times in the West heroism is regardless intertwined with martyrdom, right? So it's kind of this idea of like, you have to speak to, you know, can we always talked about justice, let justice be done though? The heavens fall. This is a common, um, kind of motto among people with conscience and that you have to do the right thing, even the consequences might not be what you like. And I think that is a good loose definition of heroism. So if you meet, I'll give you one example of heroism. This was on twitter and I really feel bad that I don't remember the guy's name. Uh, this was the line to Auschwitz, I believe it was. And you know, there's the Nazi guards keeping everyone along And uh, if you were a certain, I think if you were under 12, they killed you or some, there was some age limit where some kids were killed or something. We're not, there was some circumstances and he asked the mom how old this kid was. And she's like, he's 14 and she's like, no, he's 12. And she's like, no, he's nice, 14. She goes, he's 12. And she realized what this Nazi was telling her, even in that circumstance, and it ended up saving the kid's life. So, I think heroism in this context is defiance and standing true to values of liberalism, humanism and venerating the sanctity of human life. I think that, uh, and I think it's also important to pick your battles. I don't think if, you know, he got that Nazi over there, gotten a bullhorn and said, hey, this is the rules, blah, blah, blah, blah, blah, that's not gonna help anyone do anything. So, I do think, you know, people, a lot of times attack me for my anarchist views, it's like, oh, you know, would you call the police, would you use the roads, would you pay your income taxes? Uh, you know, I got an argument with tim pool because there was that couple, I think it was that Missouri or Illinois when they had their guns and they were being arrested and they basically took a plea deal and he said, you should have thought I go, it's a lot easier to say you should fight, but we don't know what circumstance, someone is under and what these totalitarian regimes did very, very well, as you know, is if you were a target and they can't get through to you, that's fine. You have a family. So you can sit there lex and gird your jaw and you can stand up to all the torture. Cool, what are we gonna do about your wife? What about your mom? One thing Stalin did he made it a law That kid's up to 14 and up could get the death penalty for certain crimes. So after that, the rule was from the NK VD. If you were interrogating someone, they would have death warrants for the kids child on the desk visible. So I'm interrogating you asking you to commit to, I'm sorry to admit to some crime that you're not committed and those piece of paper, it's, it's, you know, Svetlana, she's got a death warrant, you're gonna admit to any crime you want. So this is something americans, this is even the case right now in North Korea. Um, which I know you had young me park on something. I talked about a lot. Let's talk about instead of the hypothetical, but this is happening right now on earth. You can look at the map on google. The great leader kim Il Sung the founder of North Korea said class enemies must be exterminated three generations. So North, when people talk about individualism versus collectivism rick Santorum former senator says, the family is the basic unit of society unit, North Korea takes that seriously. The family is punished as a unit. So if someone does something wrong, three generations have to pay the price and you often don't know who it is that got you all in trouble. There's not a trial. This to western minds is something almost incomprehensible. [/INST] It's a lot easier to be brave when it's just your skin. There's something when it's when it's yeah, when it's your child, your your loved ones, every man becomes a coward. </s><s>[INST] But also what bravery is there for me to write an essay for The Guardian to say, I don't vote, there's no consequences to me. There's no possibility of consequences to me. This is a wonderful thing about living, excuse me in a free country. Uh it would take a lot of courage to be in the soviet union and say I'm not going to vote. And what would that courage accomplish? Very little. So, I think heroism in the sense of kind of the suicidal stuff and taking a stance with no consequences is a bit overrated. [/INST] There is some aspect, like the way I think about heroism is something that you said about the Nazi soldier, which is quietly privately in your own life, lived the virtues that you want the rest of the world to live by. </s><s>[INST] Yes, [/INST] so like without writing about it is um it's not as heroic as living it quietly. </s>

<s>[INST] I don't know if it was obvious, but it was obvious there was a significant probability that it could be a big deal. I remember it wasn't clear. There's still stories of it, the big concern, the hospital's might have been filled, but it's gonna calculating in two months [/INST] or something. Yeah. No, but there was no way we weren't going to have tens of thousands of deaths at a minimum at that point, and it was totally rational to be worried about hundreds of thousands, and when Nicholas Christakis came on my podcast very early, he predicted quite confidently that we would have about a million people dead in the US, and that didn't seem, it was, I think, appropriately hedged, but it was still, it was just like, okay, it's just gonna, you just look at the, we're just kinda riding this exponential, and we're, and it's gonna be, you know, it'd be very surprising not to have that order of magnitude, and not something much, much less. And so anyway, I mean, again, to close the story on Elon, I could see how this was being received, and I tried to get him to walk that back, and then we had a fairly long and detailed exchange on this issue, and so that intervention didn't work, and it was not done, you know, I was not an asshole, I was not, I was just concerned, for him, for the world, and then there are other relationships where I didn't take the, again, that's an example where taking the time didn't work, right, privately. There are other relationships where I thought, okay, this is just gonna be more trouble than it's worth, and I said, I just ignored it, you know? And there's a lot of that and I, Frank, again, I'm not comfortable with how this is all netted out because I don't know if, you know, frankly, I'm not comfortable with how much time in this conversation we've spent talking about these specific people. Like what good is it for me to talk about Elon or Brett </s><s>[INST] or anything? I think there's a lot of good because those friendships, listen, as a fan, these are the conversations that I loved, love as a fan and it feels like COVID has robbed the world of these conversations because you were exchanging back and forth on Twitter, but that's not what I mean by conversations, like long-form discussions, like a debate about COVID, [/INST] like a normal debate and debate. But there's no, Elon and I shouldn't be debating COVID. </s><s>[INST] You should be. Here's the thing with humility, like basically saying, we don't really know, like the Rogan method, we're just a bunch of idiots. Like one is an engineer, you're a neuroscientist, it just kind of, okay, here's the evidence and be like normal people. That's what everybody was doing. The whole world was like trying to figure out [/INST] what the hell, what? Yeah, but the issue was that at that, so at the moment I had this collision with Elon, certain things were not debatable, right? It was just, it was absolutely clear where this was going. It wasn't clear how far it was gonna go or how quickly we would mitigate it, but it was absolutely clear that it was gonna be an issue, right? The train had come off the tracks in Italy. We knew we weren't gonna seal our borders. There were already people, there are already cases known to many of us personally in the US at that point. And he was operating by a very different logic </s><s>[INST] that I couldn't engage with. Sure, but that logic represents a part of the population and there's a lot of interesting topics that have a lot of uncertainty around them, like the effectiveness of masks. [/INST] Yeah, but no, but where things broke down was not at the point of, oh, there's a lot to talk about, a lot to debate, this is all very interesting, and who knows what's what. It broke down very early at, this is, there's nothing to talk about here, like either there's a water bottle on the table or there isn't, right? </s><s>[INST] Technically, there's only one fourth of a water bottle. So what defines a water bottle? Is it the water inside the water bottle or is it the water bottle? What I'm giving you is an example [/INST] of it's worth the conversation. This is difficult because we had an exchange in private and I wanna honor not exposing the details of it, but the details convinced me that there was not a followup conversation on that topic. </s><s>[INST] On this topic. On this topic. That said, I hope, and I hope to be part of helping that happen, that the friendship was rekindled because one of the topics I care a lot about artificial intelligence, you've had great public and private conversations about this topic. And it seems like- [/INST] Yes, and Elon was very formative in my taking that issue seriously. I mean, he and I went to that initial conference in Puerto Rico together and it was only because he was going and I found out about it through him and I just wrote his coattails to it that I got dropped in that side of the pool </s><s>[INST] to hear about these concerns at that point. It would be interesting to hear how's your concern evolved with the coming out of Chad GPT and these new large language models that are fine-tuned with reinforcement learning and seemingly to be able to do some incredible human-like things. There's two questions. One, how's your concern in terms of AGI and super intelligence evolved and how impressed are you with Chad GPT [/INST] as a student of the human mind and mind in general? Well, my concern about AGI is unchanged. And so I've spoken about it a bunch on my podcast but I did a TED talk in 2016 which was the kind of summary of what that conference and various conversations I had after that did to my brain on this topic. </s>

<s>[INST] game perfect day, [/INST] Just go back after relax. Like the things that make me the happiest that tournaments is just having a good routine and feeling feeling well. Um I don't like it when too much is happening around me. So the tournament that I came from now was um the chess olympiad, which is the team land. So we were a team Norway, we did horribly. I like I did okay, but the team in general did did horribly. No, no, it'll be us, but was based on one in the end, they were this amazing team of young players. It was really, it was really impressive. But the thing is like we had a good camaraderie in the team, we had our meals together, we played a bit of football when swimming and I couldn't understand why things went wrong and I still don't understand, but the thing is for me it was all very nice, but now I'm so happy to be on my own at the tournament just to have my own routines, not see too many people otherwise just have like a very small team of people that I see </s><s>[INST] you are a kind of celebrity now. So you know, people within the chess tournament and outside would recognize you want to socialize, I want to tell you about how much you mean to them, how much you inspire them all that kind of stuff does that get in the way for you when you're like trying to really focus on on the match. Are you able to block that? Like are you able to enjoy that those little interactions and still keep your focus? [/INST] Yeah, most of the time that's uh that's fine, as long as it's not too much. But I have to admit when I'm at home in Norway, I I rarely go out with like without big headphones and something. No, not just to block out the world otherwise, um </s><s>[INST] don't make eye contact. Yeah, [/INST] no. So the thing is </s><s>[INST] people [/INST] in general are nice. I mean people, they wish me well uh and they don't like bother me also when I have the headphones on, I don't notice as much people like turning around and all of that so I can be more of in my own world. So I like that. </s><s>[INST] Yeah. What about after the, in this perfect day after the game? Do you try to analyze what happened? Do you try to think through systematically or do you just kind of loosely think about like, [/INST] no, I just loosely think about it. I've never been very structured in, in that sense. Um I know that it was always recommended that you analyze your, your own games, but I, I generally felt that I mostly had a good idea about that. Like nowadays I will like loosely see what the engine says at a certain point if I'm curious about that. Otherwise I usually move on to the next, </s><s>[INST] what about diet, you said Alma and salad and so on. I heard in your conversation with the, with the other magnus magnus number two about your, you had like this bet about meat you want, you're gonna go vegan if you lose, I forget [/INST] which vegetarian </s><s>[INST] vegetarians and uh, you both have an admiration for me. Is there um, is there some aspect about optimal performance that you look for in food? Like maybe eating only like once or twice a day or a particular kind of food? Like meat heavy diet? Is there anything like that or you just try to have fun with the food? [/INST] I think um whenever I'm tournaments, like it's very natural to eat at least for me to eat only twice a day. So usually I do that when I'm at home as well </s><s>[INST] before you do eat before, determine though you don't, you don't play fasted? [/INST] No, no. Um but I try not to eat too heavy before the game or in general to avoid sugary stuff to have a pretty stable blood sugar level because that's the easiest way to make mistake that your, your energy levels just suddenly suddenly drop and they don't necessarily mean need to be too high as long as they're pretty stable. Yeah. </s><s>[INST] Have you ever tried playing fasted? Like, you know, like intermittent fasting? So playing without having eaten? I mean the reason I asked, you know, uh especially when you do a low carb diet when I have done a person in a low carb diet, I'm able to fast for a long time, like once a day, maybe twice a day. But I just, the mind is most focused on like really difficult thinking tasks when it's fasted, it's an interesting and a lot of people kind of talk about that you're able to kind of like zoom in and if you're doing a low carb diet, you don't have this energy, the energy stable. [/INST] No, that is true. Maybe that will be interesting to try. So what's happened for me if I played a few tournaments where I've had food poisoning and then that generally means that you're both sleep deprived and you have no energy. And what I've found is that it makes me to something, it makes me very calm of course, because I don't have the energy and it makes me super creative like being slipped. It probably, I think in general makes you creative just the first thing that goes away is the ability to do the simple things that that's that's what if it affects you the most like you cannot be precise. So that's the only thing I'm worried about. Like if if I'm faster that I will won't be, won't be um precise when I play, </s><s>[INST] but you might be more creative. It's an interesting. [/INST] Yeah, potentially. </s>

<s>[INST] So in response to your eloquent description of why AI will kill us, Elon Musk replied on Twitter, okay, so what should we do about it? And you answered, the game board has already been played into a frankly awful state. There are not simple ways to throw money at the problem. If anyone comes to you with a brilliant solution like that, please, please talk to me first. I can think of things that try. They don't fit in one tweet. Two questions. One, why has the game board in your view been played into an awful state? What can just if you can give a little bit more color to the game [/INST] board and the awful state of the game board alignment is moving like this capabilities are </s><s>[INST] moving like this for the listener for the listener capability isn't moving much faster than the alignment. Yeah. All right. So just the rate of development, attention, interest, allocation of [/INST] resources, allocation of resources. We could have been working on this earlier. People are like, Oh, but you know, how can you possibly work on this earlier, because they wanted to, they didn't want to work on the problem. They wanted an excuse to wave it off, they like said, Oh, how can we possibly work on it earlier and didn't spend five minutes thinking about, is there some way to work on it earlier? Like we didn't like, and you know, frankly, it would have been hard. You know, like, like, can you post bounties for half of the physicists if your planet is taking this stuff seriously, can you post bounties for, like, half of the people wasting their lives on string theory to have gone into this instead, and try to win $1 billion with a clever solution, only if you can tell which solutions are cleverer, which is hard. We didn't take it seriously. We didn't try. It's not clear that we could have done any better if we... It's not clear how much progress we could have produced if we had tried because it is harder to produce solutions, but that doesn't mean that you're correct and justified in letting everything slide. It means that things are in a horrible state getting worse </s><s>[INST] and there's nothing you can do about it. So you're not, there's no, there's no like, there's no brain power making progress in trying to figure out how to align these systems. And you're not investing money in it, you don't have institution infrastructure for, like if you even, if you invested money in like distributing that money across the physicist or get on string theory. Brilliant minds that are working. [/INST] How can you tell through making progress? You can like put them all on interpretability cause when you have an interpretability result you can tell that it's there and there's like, but there's like, interpretability alone it's not going to save you. We need systems that will that will like have a pause button where they won't try to prevent you from pressing the pause button. Cause they were like, oh, well like I can't get it my stuff done if I'm paused. And that's like a more difficult problem. And, you know, but it's like a fairly crisp problem. You can like maybe tell somebody who's made progress on it. </s><s>[INST] So you can write and you can work on the PAWS problem. I guess more generally, the PAWS button. [/INST] More generally you can call her the control problem. I don't actually like the term control problem, because, you know, it sounds kind of controlling and alignment, not control. Like you're not trying to like take a thing that disagrees with you and like whipt it back on to like make it do what you wanted to do even though it wants to do something else you're trying to like, in the process of its creation, choose its direction. </s><s>[INST] Sure. But we currently, in a lot of the systems we designed, we do have an off switch. That's a fundamental part. [/INST] It's not smart enough to prevent you from pressing the off switch and probably not smart enough to want to prevent you from pressing the off switch. </s><s>[INST] So, you're saying the kind of systems we're talking about, even the philosophical concept [/INST] an off switch doesn't make any sense because- Well, no, the off switch makes sense. They're just not opposing your attempt to pull the off switch. Parenthetically, don't kill the system if you're... If we're getting to the part where this starts to actually matter and where they can fight back, don't kill them and dump their memory. Save them to disk. Don't kill them. </s><s>[INST] Be nice here. Well, okay, be nice is a very interesting concept here. We're talking about a system that can do a lot of damage. I don't know if it's possible, but it's certainly one of the things you could try is to have an off switch. A suspend to disk switch. You have this romantic attachment to the code. Yes, if that makes sense. But if it's spreading, [/INST] you don't want suspend to disk. There's something fundamentally broken. </s><s>[INST] If it gets that far out of hand, then yes, pull the plugin and everything is running on. Yes. I think it's a research question. Is it possible in AGI systems, AI systems, to have a sufficiently robust off switch that cannot be manipulated? That cannot be manipulated by the AI system? [/INST] Then it escapes from whichever system you've built the almighty lever into and copies itself </s><s>[INST] somewhere else. Your answer to that research question is no, but I don't know if that's [/INST] a hundred percent answer. I don't know if it's obvious. I think you're not putting yourself </s>

<s>[INST] does that feel? The fact that you're probably somewhat alone, maybe a couple of friends recording with the logitech webcam and kind of going home alone at one or two a.m. At night and knowing that that's gonna reach sort of thousands of people, eventually millions of people is what's that feeling like? I mean is there a feeling of just satisfaction of pushing through? [/INST] I think it's humbling and I wasn't thinking about what I was feeling. I think one thing we uh I'm proud to say we got right from the early days was um I told my whole team back then that the number one priority is to do what's best for learners do, what's best for students. And so when I went to the recording studio, the only thing on my mind was what can I say, how can I design my size where you need to draw right to make these concepts as clear as possible for learners? Um I think, you know, I've seen sometimes instructors is tempting to, hey, let's talk about my work, maybe if I teach you about my research, someone will cite my papers a couple more times. And I think one things we got very launched the first few months and later building called Sarah was putting in place that bedrock principle of let's just do what's best for learners and forget about everything else. And I think that that is a guiding principle turns out to be really important to the to the rise of the movement </s><s>[INST] and the kind of learner you imagined in your mind is as as broad as possible, as global as possible. So really try to reach as many people interested in machine learning and ai as possible. [/INST] I really want to help anyone that had an interest in machine learning uh to break into the field. And and I think sometimes eventually people ask me, hey, why are you spending so much time explaining great and dissent? And and and my answer was um if I look at what I think to learn they need someone benefit from. I felt that having that a good understanding of the foundations, coming back to the basics would put them in a better state to then build on a long term career, so try to consistently make decisions on that principle. </s><s>[INST] So one of the things you actually revealed to the narrow Ai community at the time and to the world is that the amount of people who are actually interested in ai is much larger than we imagined. Buy you teaching the class and how popular it became. It showed that, wow, this isn't just a small community of sort of people who go to in your apps and and it's it's much bigger. It's developers, it's people from all over the world. I mean, I'm Russian, so everybody in Russia is really interesting. This is a huge number of programmers who are interested in machine learning, India china, uh South America everywhere that there's just millions of people who are interested in machine learning. So how big do you get a sense that this is the number of people is that are interested from your perspective? [/INST] I think the numbers grown over time. I think a lot of those things that maybe it feels like it came out of nowhere, but it's an inside the building, it it took years. It's one of those overnight successes that took years to get there. My first foray into this type of education was when we're filming my standard clause and sticking the videos on YouTube and then some of the things we uploaded the whole works and so on. But you know, basically the one hour, 15 minute video that we put on YouTube. Um and then we had four or five other versions of websites that have built uh, most of what you would never have heard of because they reach small audiences. But that allowed me to reiterate, allow my team and me to elevate to learn what the ideas that work and what doesn't. For example, one of the features I was really excited about and really proud of was built this website where multiple people could be logged into the website at the same time. So today if you go to a website, you know, if you are logged in and then I want to log in you need to log out. This is the same browsing the same computer. But I thought, well what if two people say you and me, we're watching a video together in front of a computer. What if a website could have, you type your name and password, have me type my name, password and then now the computer knows both of us are walking together and it gives both of us credit for anything we do as a group Uh, influences feature rolled it out in a high, in a school in San Francisco we had about 20 something users um, where's the teacher, their sacred heart, cathedral prep, the teacher is great. Um, and guess what zero people use this feature. Uh, it turns out people studying online. They want to watch the videos by themselves, you can playback, pause at your own speed rather than in groups. So that was one example of a tiny lesson learned all the many that allows us to hone into the set of features and </s>

<s>[INST] So, another fascinating place um in a paper title pushing the frontiers of density functional by solving the fractional electron problem. So, you're taking on modeling and simulating the quantum mechanical behavior of electrons. Can you explain this work and can ai model and simulate arbitrary quantum mechanical systems in the future? Yeah, [/INST] so this is another problem I've had my eye on for, you know, a decade or more, which is um sort of simulating the properties of electrons. If you can do that, you can basically describe how elements and materials and substances work. So it's kind of like fundamental if you wanna advance material science, um and, you know, we have Schrodinger's equation and then we have approximations to that density functional theory. These things are, you know, are famous and people try and write approximations to to these, to these functional and and kind of come up with descriptions of the electron clouds where they're gonna go, how they're gonna interact when you put two elements together and what we try to do is learn a simulation, learn a functional that will describe more chemistry types of chemistry. So, until now, you know, you can run expensive simulations, but then you can only simulate very small molecules, very simple molecules. We would like to simulate large materials. Um and so today there's no way of doing that. And we're building up towards building functional. Is that approximate schrodinger's equation. And then allow you to describe what the electrons are doing. Um And all materials sort of science and material properties are governed by the electrons and and how they interact. </s><s>[INST] So, have a good summarization of the simulation through the functional um but one that is still close to what the actual simulation would come out with. So what mm, how difficult that task. What's involved in that task is running those, those complicated simulations and learning the task of mapping from the initial conditions and the parameters of the simulation, learning what the functional would [/INST] be. So it's pretty tricky. And we've done it with you know the nice thing is we there are we can run a lot of the simulations. The molecular dynamics simulations on our compute clusters and so that generates a lot of data. So in this case the data is generated. So we like those sort of systems and that's why we use games Simulator generated data and we can kind of create as much of it as we want really. Um And just let's leave some you know if any computers are free in the cloud we just run we run some of these calculations right compute cluster calculations. </s><s>[INST] The free computer time is used up on quantum mechanics. [/INST] Quantum mechanics exactly simulations and protein simulations and other things. And so um and so you know when you're not searching on Youtube for video cat videos, we're using those computers usefully And Quantum chemistry the idea and putting them to good use and then Yeah. And then all of that computational data that's generated we can then try and learn the functional from that. Which of course are way more efficient. Once we learn the functional than running those simulations would be </s><s>[INST] do you think one day aI may allow us to do something like basically crack open physics so do something like travel faster than the speed of light. [/INST] My ultimate aim has always been with Ai is um the reason I am personally working on Ai for my whole life was to build a tool to help us understand stand the universe. So I wanted to and that means physics really and the nature of reality. So um I don't think we have systems that are capable of doing that yet, but when we get towards A G I I think that's one of the first things I think we should apply A G. I to. I would like to test the limits of physics and our knowledge of physics. There's so many things we don't know. There's one thing I find fascinating about science and you know, it's a huge proponent of the scientific method as being one of the greatest ideas humanity has ever had and allowed us to progress with our knowledge. I think as a true scientist, I think what you find is the more you find out the more you realize we don't know. And and I always think that it's surprising that more people don't aren't troubled, you know, every night. I think about all these things we interact with all the time that we have no idea how they work time, consciousness, gravity, life we can't I mean these are all the fundamental things of nature. </s><s>[INST] I think the way we [/INST] don't really know what they are </s><s>[INST] to live life. We pin certain assumptions on them and kind of treat our assumptions as if they're a fact that allows us to sort of [/INST] box them off somehow. </s><s>[INST] Box them off. But the reality is when you think of time, you should remind yourself, you should put it off, take it off the shelf and realize like no, we have a bunch of assumptions. There's still a lot of there's even not a lot of debate. There's a lot of uncertainty about exactly what is time. Uh is there an era of time? You know, there's there's a lot of fun about the questions that you can't just make assumptions about and maybe ai allows you to um not put anything on the shelf. Yeah, not make any hard assumptions and really open it up and see what [/INST] exactly. I think we should be truly open minded about that. And exactly that not be dogmatic to a particular theory. Um It'll also allow us to build better tools, experimental tools eventually, that can then test certain theories that may not be testable today about things about like what we spoke about the beginning about the computational nature of the universe. How one might if that was true. How one might go about testing that, right? And and how much, you know, there are people who have conjectured people like Scott Aaronson and others about, you know, how much information can a specific plank unit of space and time contain. Right? So one might be able to think about testing those ideas if you had um Ai helping you build some new, exquisite experimental tools. This is what I imagine that you know many decades from now we'll be able to do </s>

<s>[INST] They sort of basically match the resumes to the job description and give you a really nice initial pool and then help you flip through that pool down. I mean the whole process from the beginning pool to the end perfect candidate that will make your life a beautiful flourishing experience. They help you out with that. Indeed, has a special offer only available for a limited time. Check it out at indeed dot com slash lex. This is the lex Friedman podcast. To support it. Please check out our sponsors in the description and now dear friends, here's Ray Kurzweil in your 2005 book titled The singularity is near you predicted that the singularity will happen in 2045. So now, 18 years later, do you still estimate that the singularity will happen on 2045 maybe first, what is the singularity? The technological singularity? And when will it happen? [/INST] Singularity is what computers really change? Our view of what's important and change who we are. But we're getting close to some salient things that will change who we are. The key thing is 2029 when computers will pass the turing test and there's also some controversy whether the turing test is valid. I believe it is. Uh Most people do believe that but there's some controversy about that. But stanford got very alarmed at my prediction about 2029 I made this in 1999 in my book the </s><s>[INST] age of spiritual machines and then you repeated the prediction in 2005, [/INST] 2000 </s><s>[INST] five. So [/INST] they held international conference, you might have been aware of it of Ai experts in 1999 to assess this view. So people gave different predictions and they took a poll was really the first time that Ai experts worldwide were polled on this prediction. Uh and the average poll was 100 years, 20% believed it would never happen. And that was the view in 1999 </s><s>[INST] 80 [/INST] percent believed it would happen but not within their lifetimes. There's been so many advances in ai uh that the poll of ai experts has come down over the years. So a year ago something called Meticulous which may be aware of assesses different types of experts on the future. They again assessed what Ai experts then felt and they were saying 2042 for the turning test. </s><s>[INST] So it's coming down [/INST] and I was still saying 2029 a few weeks ago again did another poll and it was 2030. So a experts now basically agree with me, I haven't changed at all, I've stayed with 2029. Um and the experts now agree with me but they didn't agree at first. </s><s>[INST] So alan turing formulated the turing test and [/INST] right now what he said was very little about it. I mean the 1950 paper where he had articulated the turing test, he is like a few lines that uh talk about the turing test um and it really wasn't very clear how to administer it and and he said if if they did it in like 15 minutes </s><s>[INST] that [/INST] would be sufficient, which I don't really think is the case. These large language models. Now. Some people are convinced by it already, I mean you can talk to it and have a conversation with you and you can actually talk to it for hours. Um So it requires a little more depth. There's some problems with large language models which we can talk about. Um But some people are convinced by the turing test. Now, if somebody passes the turing test, what what are the implications of that? Does that mean that they're sentient? They're conscious or not, it's not necessarily clear what the implications are. Anyway, I believe 2029 that's 67 years from now. Uh We'll have something that passes the turing test and a valid during test meaning it goes for hours, not just a few minutes, </s><s>[INST] please speak to that a little bit. What is your formulation of the touring test? You've proposed a very difficult version of the turing test. So what what does that look like [/INST] basically? It's just to assess it over several hours. Um and also have a human judge that's fairly sophisticated on what computers can do and can't do Um If you take somebody who's not that sophisticated or even a average engineer, uh they may not really assess various aspects of it. So </s><s>[INST] you really want the human to challenge the system [/INST] exactly, exactly </s><s>[INST] on its ability to do things like common sense reasoning. Perhaps [/INST] that's actually a key problem with large language models. They don't do uh these kinds of um tests that would involve assessing chains of reasoning. Um but you can lose track of that. If you if you talk to them, they actually can talk to you pretty well and you can be convinced by it, but it's somebody that would really convince you that it's a human, uh whatever that takes, maybe it would take days or weeks. Um but it would really convince you that it's human. Um Large language models uh can appear that way. You can read conversations and that they appear pretty good. There are some problems with it. It doesn't do math very well. You can ask how many legs did 10 elephants have, and they'll tell you, well, okay, each elephant has four legs and 10 elephants. So it's 40 legs. And you go, okay, that's pretty good. How many legs do 11 elephants have? And they don't seem to understand the question. Do </s><s>[INST] all humans understand that question? No, that's [/INST] the key thing. I mean how advanced the human do you want it to be, But we do we do expect a human to be able to do multi chain reasoning to be able to take a few facts and put them together. Not perfectly. We see that you know in a lot of polls that people don't do that perfectly at all. But um so it's not it's not very well defined but it's it's something where it really would convince you that it's human </s>

<s>[INST] right now in that that process, you know, with artificial neural networks, it's a very simple process back propagation that adjust the weights, the process of synaptic genesis genesis Chapter genesis it's even easier. It's even easier. [/INST] It's even easier. Back propagation requires something we uh it really can't happen in brains. This back propagation of this area. Signal that it really can't happen. People are trying to make it happen in brains. That doesn't happen. This is this is pure heavy and learning Well, it's not the genesis is pure heavy and learning. It's basically saying there's a population of cells over here that are active right now and there's a population of cells over here active right now, how do I form connections between those active cells And it's literally saying this guy became active, this these 100 neurons here became active before this neuron became active. So form connections to those ones. That's it. There's no propagation of error, nothing. All the networks, we do all models, we have work on almost completely on heavy and learning, but in in on dendritic segments and multiple synapses at the same time. </s><s>[INST] So now let's have turned the question that you already answered and maybe you can answer it again, if you look at the history of artificial intelligence, where do you think we stand? How far are we from solving intelligence? You said you were very optimistic. Can you elaborate on that? [/INST] Yeah. You know, it's always the crazy question to ask Because you know, no one can predict the future. So I'll tell you a story. I used to uh I used to run a different neuroscience institute called the redwood neuroscience to two and we would we would hold these symposiums and we got like 35 scientists around the world to come together. And I used to ask them all the same question, I would say well how long do you think it'll be before we we understand how the new york cortex works And everyone went around the room and they had introduced the name and they have to answer that question. So I got The typical answer was 50 200 years. Some people would say 500 years, some people said never. I said, what are you while you're a neuroscientist, it's good pay. Um it's interesting. Um so, you know, but it doesn't work like that, as I mentioned earlier, these are not, these are step functions, things happen and then bingo that happened. You can't predict that. Um I feel I've already passed that step function. So if I can do my job correctly over the next five years, um then meaning I can proselytize these ideas, I can convince other people that right? We can show that other people, uh machine learning people should pay attention to these ideas, then we're definitely in an under 20 year time frame, if I can do those things, if I if I'm not successful in that, and this is the last time anyone talks to me and no one reads our papers and you know, I'm wrong or something like that, then um then I don't know, But it's it's not 50 years, it's it, You know, it'll it'll, you know, the same thing about electric cars, how quickly are they going to populate the world? It's probably takes about a 20 year span. Um it'll be something like that, but I think if I can do what I said, we're starting it </s><s>[INST] and of course there could be other use of step functions, it could be, Everybody gives up on your ideas for 20 years and then all of a sudden somebody picks it up again, wait, that guy was onto [/INST] Something. So that would be that would be a failure on my part. Right. You know, think about Charles Babbage, you know, Charles Babbage. This guy has to have invented the computer back in 18 something, 1800s. Everyone forgot about it until, you know, 100 years later. Hey, this guy figured this stuff out a long time ago, you know, but he was ahead of his time, I don't think, you know, like, uh, as I said, uh, I recognize this is part of any entrepreneurs challenge and I use the entrepreneur broadly in this case, I'm not meaning like I'm building a business trying to sell something. I mean, I'm trying to sell ideas and this is the challenge as to how you get people to pay attention to you, How do you get them to give you positive or negative feedback? Um, how do you get people act differently based on your ideas? So, um, you know, we'll see how, what we do on them. </s><s>[INST] So, you know, that there's a lot of hype behind artificial intelligence currently, do you? Uh, as as you look to spread the ideas that are of no cortical theory, the things you're working on, Do you think there is some possibility will hit an Ai winter? Once again, [/INST] it's certainly it's certainly a possibility that I worry about. Yeah, well I guess do I worry about it. Um I haven't decided yet if that's good or bad for my mission. </s><s>[INST] That's true. That's very true because it's almost like you need the winter to refresh the palate. [/INST] Yeah. So it's like I want here's what you want to have. It is you want like to extend that. Everyone is so thrilled about the current state of machine learning and Ai and they don't imagine they need anything else that makes my job harder if if everything crashed completely and every student left the field and there was no money for anybody to do anything. And it became an embarrassment to talk about machine intelligence and Ai. That wouldn't be good for us either. You want you want sort of the soft landing approach, right? You want enough people in the senior people in AI and machine learning and say, you know, we need other approaches. We really need other approaches. But damn we need other approaches. Maybe we should look to the brain. Okay, let's look who's got some brain ideas. Okay, let's let's start a little project on the side here trying to do brain idea related stuff. That's the ideal outcome we would want. So I don't want a total winter and yet I don't want it to be sunny all the time </s>

<s>[INST] So just for people to not familiar, I mean the way you detect aircraft. So I mean I'm sure there's a lot of ways, but radar for the longest time, there's a big blob that appears in the radar. How do you make a plane disappear? So it looks as big as a ball bearing. What's involved in technology wise there, what's uh broadly sort of the stuff you can [/INST] speak about. They'll stick to what's in Ben Rich's book, but obviously the geometry of how radar gets reflected and the kinds of materials that either reflect or absorb or some kind of a couple of the critical elements there. Um and it's a cat and mouse game. Right? I mean, you know, radars get better stealth capabilities get better. And so it's a it's a really uh game of continuous improvement and innovation there. I'll leave it at that. </s><s>[INST] Yeah. So the idea that something is essentially invisible Is quite fascinating, but the other one is flying fast. The speed of sound is 750, mph. Uh, so supersonic is three, you know Mach three, something like that. [/INST] When we talk about the supersonic obviously and we kind of talked about that is that realm from Mach one up through about Mach five and then hypersonic so you know Hi Supersonic Speeds would be passed Mach five. And you gotta remember Lockheed martin and actually other companies have been involved in hypersonic development since the late sixties. Um You know you think of everything from the X 15 to the space shuttle as examples of that. I think the difference now is uh if you look around the world particularly the threat environment that we're in today you're starting to see you know publicly uh folks like the Russians and the chinese um saying they have hypersonic weapons capability that could threaten uh U. S. And allied capabilities. And also basically you know the claims are these could get around defensive systems that are out there today And so there's a real sense of urgency. You hear it from folks like the undersecretary of defense for research and engineering. Dr mike Griffin and others in the Department of Defense that hypersonic is uh is something that's really important to the nation in terms of both parody but also defensive capabilities. And so that's something that we're pleased. It's something Lockheed martin's had a heritage in. We've invested R. And D. Dollars on our side for many years. And uh we have a number of things going on with various us government customers in that field today that we're very excited about. So I would anticipate we'll be hearing more about that in the future from our customers. </s><s>[INST] And I actually haven't read much about this. Probably you can't talk about much of it at all. But on the defensive side is a fascinating problem of perception of trying to detect things that are really hard to see. Can [/INST] you comment on how </s><s>[INST] hard that problem is and uh how how hard is it to stay ahead? Uh even if we're going back a few decades, stay ahead of [/INST] the competition? Well, maybe maybe I again you got to think of these is ongoing capability development. And uh so think back to the early days of missile defense. So this would be in the eighties the S. D. I. Program. And in that time frame we proved Lockheed martin proved that you could hit a bullet with a bullet essentially. And uh which is something that had never been done before to take out an incoming ballistic missile. And so that's led to these incredible hit. To kill kinds of capabilities. Pack three. Um that's the Patriot advanced capability uh model three. That Lockheed martin builds the thaad system that I that I talked about. Um so now hypersonic uh you know, they're different from ballistic systems. And so we got to take the next step in defensive capability. </s><s>[INST] I can I'll leave that there. But I can only imagine. No let me just comment sort of as an engineer. It's sad to know that so much that locket is done in the past is classified or today, you know and it's shrouded in secrecy. It has to be by the nature of the application. So like what I do so what we do here at M. I. T. Would like to inspire young engineers, young scientists and yet in the lucky case some of that engineer has to stay quiet. How do you think about that? How does that make you feel? Is there a future where more can be shown? Or is it just the nature the nature of this world that it has to remain secret? [/INST] It's a good question. I think the public can see enough of and including students who may be in grade school, high school college today. Uh to understand the kinds of really hard problems that we work on And I mean look at the F- 35, right? And uh you know obviously a lot of the detailed performance levels are are sensitive and control. But you know we can talk about what an incredible aircraft this is. You know, supersonic, super cruise kind of a fighter, a a you know stealth capabilities, its a flying information, you know, system in the sky with data fusion sensor fusion capabilities that have never been seen before. So these are the kinds of things that I believe, you know that these are kinds of things that got me excited when I was a student. I think these still inspire students today. And the other thing and say, I mean, you know, people are inspired by space, people are inspired by aircraft. Our employees are also inspired by that sense of mission and I'll just give you an example. I had the privilege to work and and lead our GPS programs for some time. And uh that was a case where you actually worked on a program that touches billions of people every day. And so when I said, I worked on GPS, everybody knew what I was talking about, even though they didn't maybe appreciate the technical challenges that went into that. But I'll tell you got a briefing one time from a major in the Air force and he said, I I go by call sign Gimp Gps is my passion. I love GPS and he was involved in the operational test of the system. He said, I went, I was out um in Iraq and I was on a helicopter, uh Black hawk helicopter um uh it was bringing back, you know sergeant and a handful of troops from a deployed location. And uh you know, he said my job is GPS. So I asked that sergeant, he's you know beaten down and kind of half asleep. And I said, what do you think about GPS? And he brightened up, his eyes lit up and he said, well, GPS that brings me and my troops home every day. I love GPS. And that's the kind of story where it's like, okay, I'm really making a difference here in the kind of work. So that that mission piece is really important. The last thing I'll say is, and this gets to some of these questions around advanced technologies. It's not, you know, they're not just airplanes and spacecraft anymore for people who are excited about advanced software capabilities, about ai about bringing machine. These are the uh, these are the things that we're doing to exponentially increase the mission capabilities that go on those platforms and those are the kinds of things I think are more and more visible to the public. </s>

<s>[INST] Okay, so the building engineering, the Neocortex in itself is just creating an intelligent prediction system, modeling system, sorry, modeling system. Uh you can use it to make predictions. And um but you can also put it inside a thing that's actually acting in this world, [/INST] you have to put it inside something. It's again, think of the map analogy right? Map on its own doesn't do anything right, It's just an earth, it's just that you can learn, but it's just so we have to embed it somehow and something to do something. </s><s>[INST] So so what's your intuition here? You had, you had a conversation with Sam Harris recently that was sort of um you've had a bit of a disagreement and you're sticking on this point, you know, Elon musk Stuart, russell kinda have a worry, existential threats of ai what's your intuition? Why if we engineer increasingly intelligent neocortex type of system in the computer, why that shouldn't be a thing that we [/INST] it was used the word intuition and Sam Harris used the word intuition to and and we didn't use that intuition that were immediately stopped and said that's the crux of the problem. He's using intuition. I'm not speaking about my intuition, I'm speaking with something, I understand something, I'm going to build something. I am building something. I understand completely or at least well enough to know what it's, I'm guessing I know what this thing is going to do and I think most people who are worried they have trouble separating out, they don't have, they don't have the knowledge or the understanding about like what is intelligence, how is it manifested the brain? How's it separate from these other functions in the brain. And so they imagine there's going to be human like or animal like it's going to have, it's gonna have the same sort of drives and emotions we have but there's no reason for that, that's just because there's there's unknown if you if the unknown is like oh my God you know I don't know what this is gonna do, we have to be careful, it could be like us but really smarter. I'm saying you know it won't be like us, it would be really smart but it won't be like us at all. And um and but I'm coming from that not because I just guessing I'm not into it using intuition I'm basically like ok I understand this thing works, that's what it does explain it to you </s><s>[INST] okay? But to push back so I I also disagree with the intuitions that SAM has but but I disagree with what you just said, which you know what's a good analogy. So if you look at the twitter algorithm in the early days just recommend our systems. You can understand how recommend our systems work. What you can't understand in the early days is when you apply that recommended system at scale to thousands and millions of people how that can change societies. So the question is yes you're just saying this is how an engineering neocortex works but like when you have a very useful ah tick tock type of service that goes viral when your neocortex goes viral and then millions of people start using it cannot destroy the world. [/INST] Uh Well first of all this is back. One thing I want to say is that A i is a dangerous technology. I'm not denying that. </s><s>[INST] All technologies dangerous. [/INST] Well and ai may be particularly so. Okay, so um am I worried about it? Yeah, I'm totally worried about it. The thing where the narrow component we're talking about now is the existential risk of a right? So I want to make that distinction because I think a I can be applied poorly. It can be applied in ways that you know, people are going to understand the consequences of it. Um These are all potentially very bad things but they're not the ai system creating this existential risk on its own. And that's the only place I disagree with other people. </s><s>[INST] So I I think the existential risk things, humans are really damn good at surviving. So to kill off the human race be very very you can [/INST] even Yes, but you can even I'll go further. I don't think Ai systems are ever going to try to I don't think I systems are ever going to like say I'm gonna ignore you, I'm gonna do what I think is best. Um I don't think that's going to happen at least not in the way I'm talking about it. So you the twitter recommendation, Outrun this interesting example, Let's use computers and analogy again, right? I build a computer, it's a universal computing machine. I can't predict what people are going to use it for. They can build all kinds of things. They can they can even create computer viruses, it's, you know, all kinds of stuff. So there's some unknown about its utility, about where it's going to go. But on the other hand, I pointed out that once I build a computer it's not going to fundamentally change how it computes. Its like I use the example of a register which is a part internal part of a computer. Um you know, I say it can't just because computers don't involve, they don't replicate, they don't involve, they don't, you know, the physical manifested computer itself is not gonna there's certain things that can't do, right? So we can break into things like things that are possible to happen, we can't predict and things are just impossible to happen unless we go out of our way to make them happen. They're not going to happen unless somebody makes them happen. </s>

<s>[INST] he [/INST] asked his question, he says can you imagine to one day work for the government? And so I gave a pretty clever answer. I said yes but not as a chemist. So I answered the question that he didn't ask, I helped him </s><s>[INST] out. [/INST] So we made an arrangement to meet for lunch, which in Germany is the main meal at the number one restaurant in indiana, you know, I still remember what I ate </s><s>[INST] uh rump [/INST] steak with with butter on top and french fries was my favorite. Anyway, um so when I get to the restaurant and I saw this fellow sitting in the back there at the table and uh there was another person at the table. So I was a little bit hesitant because in those days it was not unusual for perfect strangers to share a table because there wasn't wasn't enough tables and chairs and so forth. So I didn't know if I could approach him, but he got up and came to me and he took me to the table and he said I want to introduce uh Herman. We worked with our soviet comrades. Ha ha KGB and then he he disappeared. He says I got something else to do, I never knew his name. He just handed me over to the KGB. </s><s>[INST] What was the relationship between the KGB and the Stasi collaborators? Close collaborators? Or just distant associates? [/INST] They were pretty close collaborators as I told you that you know, they bought forged documents that the Germans made because the Germans were better at forgery. They also exchanged information but they didn't trust each other. 100%. And and I tell you why I know that. So they recruited me to send me to West Germany. As I already said East Germany had 1000 agents over there. Why would they have they want to have their own </s><s>[INST] Yeah, yeah. Okay, this is fascinating. Internal and external dynamic of distrust. Okay, so there you are welcomed by the KGB. When did the offer the invite come? [/INST] Well that took a while. So Herman and I had an unofficial relationship for about a year and a half. I would meet him once a week, once every two weeks initially in his car. But then he, he took me to a const operational flat was an apartment that was occupied by a party member, a lady, single lady. When we came in, she would leave, she left us tea and cookies and then we could freely talk. He also at that time gave me some West german literature magazines to read which was of course forbidden. So already I'm starting to feel somewhat special and as we were talking about what they had in mind for me in general, I knew that I wasn't going to be even more special because I would be above the law. I would, I would operate outside the law of the countries I would go to as well as East Germany because you know the magazines and uh and eventually when, when I joined up, they told me I had better watch West german television which was also not explicitly prohibited, but it was uh something that could get you in trouble. So </s><s>[INST] on many levels you're super special, you're the Jazz. [/INST] Yes, yes. </s><s>[INST] So what was that recruitment testing process? Like testing whether you are you have what it takes to be a [/INST] KGB agent. First of all, um we had a very in depth talks uh Herman and I about life and I I was I still am very honest and sharing my feelings </s><s>[INST] philosophical or personal, [/INST] personal. I even I even told him that I was shy around the girls. Uh </s><s>[INST] he was he was giving you relationship advice or [/INST] what </s><s>[INST] was so what was the dynamic? Can you tell me? Was it a father son? [/INST] Yeah, he was maybe in this uh Early to mid 30s and I was maybe 10 years younger. </s><s>[INST] And what languages did he speak, [/INST] Speak German? He spoke german pretty well, but </s><s>[INST] he's originally from [/INST] Yeah, with a Russian accent. Uh so I got in trouble one time with him when when I asked him is your real name German, he didn't like that, </s><s>[INST] He didn't like it. Was he good with girls? Was was No, no, he just, [/INST] you know, I remember what he told me, he says, you know, you've got to understand one thing they're looking for guys too. That's that's all uh </s><s>[INST] girls are looking at. It's a competitive game. [/INST] Yeah, don't don't don't don't worry about it. You know, don't be so shy. </s><s>[INST] So that little flame of love that we talked about in all the shapes that it takes in our life, did he talk to you about that? That that could be taken advantage of that, that could be used or was it implied. Yeah, [/INST] but not in it was not very focused, not in great detail. So let's so we talked about personal stuff and you know, like dislikes him. He gave me tasks for instance, when my friend and I hitchhiked from from east Germany all the way down to Bulgaria. He told me to write a report about it, what I saw. So fundamentally, he wanted to see how well I can I can write and how well I can report how well I observe. Uh he also asked me to write some profiles about fellow students. I don't believe that was for them to give them to the Stasi it was just like how well do I characterize people that's important when you're talking about when I was in the U. S. Active in the U. S. I operated as a spotter. So I did exactly that I wrote profiles about people. Ah he also gave me some tasks to do that. We're rather unpleasant. Um uh he would give me an address in the name of uh of uh the people who lived at the address. And he told me to go there, ring the doorbell and find out something about a relative who lived in West Germany that is undercover exploration. Right? So you go you you make up a story and somehow win the confidence of your target to tell you something that you want to know, </s>

<s>[INST] It's very intuitive, it's very simple. It does the thing it's supposed to do and does it well in terms of execution, there's very few things I admire then systems, software, systems that do the job they're supposed to do and do it well. And all the improvements, all the version updates that they do. Just push further the ability to do the thing. Well they don't shove all the kind of weird features into it. It just does the thing is supposed to do. And does it well go to express VPN dot com slash body for an extra three months free. This is a lex Friedman podcast to support it, please check out our sponsors in the description and now dear friends, here's live very what role do you think luck plays in poker and in life you can pick whichever one you want poker or life and or life, [/INST] the longer you play, the less influenced luck has, you know, like with all things, the bigger your sample size, the more the quality of your decisions or your strategies matter. Um, so to answer that question you in poker, it really depends if you and I sat and played 10 hands right now, I might only win 52% of the time, 53% maybe. Um, but if we play 10,000 hands then I'll probably win like over 98 99% of the time. So it's a question of sample sizes </s><s>[INST] and what are you figuring out over time, the betting strategy that this individual does or literally doesn't matter against any individual overtime [/INST] against any individual over time, the better player because they're making better decisions. So what does that mean to make a better decision? Well, uh to get into the real nitty gritty already um basically poker is a game of math. Um there are these strategies familiar with like nash equilibrium. That's right. So there are these game theory, optimal strategies that you can adopt. Um and the closer you play to them, the less exploitable you are. So because I've studied the game a bunch, although admittedly not for a few years but back in, you know, when I was playing all the time, um I would study these game theory optimal solutions and try and then adopt those strategies when I go and play. So I play against you and I would do that. And because the objective, when you're playing game theory optimal, it's actually it's a loss minimization thing that you're trying to do. Um your best bet is to try and play uh sort of similar style. You also need to try and adopt this loss minimization. Um But because I've been playing much longer than you, I'll be better at that. So first of all you're not taking advantage of my mistakes, but then on top of that, I'll be better at recognizing when you are playing sub optimally and then deviating from this game theory optimal strategy to exploit your bad plays. </s><s>[INST] Can you define game theory and nash equilibrium? Can we try to sneak up to it in a bunch of ways like what's the game theory framework of analyzing poker, analyzing any kind of situation? [/INST] So game theory is just basically the study of decisions within a competitive situation. Um I mean technically a branch of economics. Um But it also applies to a wider decision theory. Um And you know, usually when you see it, it's these little payoff matrices and so on. That's how it's depicted. But it's essentially just like the study of strategies under different competitive situations. Um and as it happens, certain games, in fact, many, many games have these things called nash equilibrium. What that means is when you're in a nash equilibrium, basically, uh it is not, there is no strategy that you can take that would be more beneficial than the one you are currently taking assuming your opponent is also doing the same thing. Um So it would be a bad idea, you know, if we're both playing uh you know, a game theory, optimal strategy, if either of us deviate from that. Now, the, you know, the we're putting ourselves at a disadvantage. Um Rock paper says this is actually really great example of this. Like if we Two were to start playing rock paper, you know, you know, nothing about me and we're gonna play for all our money, let's play 10 rounds of it. What would your sort of optimal strategy be do you think? What would you do? </s><s>[INST] Um Let's see, I would probably try to be as random as possible. [/INST] Exactly, you want to, because you don't know anything about me, you don't want to give anything about away about yourself. So ideally you'd have like a little dice or somewhat, you know, perfect randomizer that makes you randomize 33% of the time each of the three different things. And in response to that. Um well actually I can kind of do anything, but I would probably just randomize back to, but actually it wouldn't matter because I know that you're playing randomly. Um So that would be us international equilibrium where we're both playing this like unexplainable strategy. However, if after a while you then notice that I'm playing rock a little bit more often than I should. </s><s>[INST] Yeah, you're the kind of person that would do that, wouldn't you? [/INST] Sure? Yes, Yes, yes, I'm more of a sister's girl. But </s><s>[INST] anyway, you [/INST] know, I'm, as I said, randomizer, so you notice I'm throwing rock too much or something like that. Now you'd be making a mistake by continuing playing this game theory, optimal strategy because the previous one, because you are now there's a I'm making a mistake and you're not deviating and exploiting my mistake. Um So you'd want to start throwing paper a bit more often um in whatever you figure is the right sort of percentage of the time that I'm throwing rock too often. So that's basically an example of where you know what what game theory optimal strategy is in terms of lost minimization, but it's not always the maximally profitable thing if your opponent is doing stupid stupid stuff, which, you know, in that example, so that's kind of how it works in poker, but it's a lot more complex. Um And the way poker players typically, you know, nowadays they study the game's changed so much and I think we should talk about how it sort of evolved. Um But nowadays like the top pros basically spend all their time in between sessions running these simulators using like software where they do basically monte Carlo simulations sort of doing billions of fictitious self play hands. You input a fictitious hand scenario, like what do I do with Jack nine suited on a king 10 for 22 spades board. Um uh and you know, against this bet size, so you'd input that, press play. It'll run, it's it's, you know, it's billions of fake hands and then it will converge upon what the game theory, optimal strategies are. Um And then you want to try and memorize what these are basically, they're like ratios of how often, you know, what types of hands you want to bluff and what percentage of the time. So then there's this additional layer of inbuilt randomization built in. Yeah, </s>

<s>[INST] starting to like [/INST] stuff just like, just hold on a second, but there will always be a path for the institutional, that will always be there for you. But now there's something else now, there's another game in town and that's more appealing to millions and millions and millions and millions of people who feel unserved by the corporate media CNN and these people possibly who feel unserved um in the, you know, the faculty, like if you are an up and comer who wants to teach as many young people as possible, I think you should be on Youtube. Look at the khan academy, God that God created a huge business. So I just think we can be cynical and upset about what that system is, but we should also have hope. Like I have a lot of hope for what can be in the future. </s><s>[INST] Yeah, there's a, there's a guy people should check out my story is a little bit different because I basically uh stepped aside with the dream of being an entrepreneur earlier in the pipeline than like a, like a legitimate, like senior faculty would, there's an example of somebody that people should check Andrew Huberman from stanford who's a neuroscientist Who's as world class as it gets in terms of like 10ured faculty, just a really world class researcher and now he's doing youtube [/INST] instagram </s><s>[INST] and so he switched. So he not just does instagram, he now has a podcast and he's doing, he's changing the nature of like I believe that Andrew might be the future of stanford and for a lot. It's funny, like he's basically joe Rogan is an inspiration to Andrew And and to me as well, and those ripple effects and Andrew is an inspiration probably just like you're saying to these young, like 25 year olds who are soon to become faculty, if we're just talking about academia and the same is probably happening with government is funny enough, trump probably is inspiring a huge number of people who are saying, wait a minute, I don't have to play by the rules. And I have to I can think outside the box here and you're right, and the institutions we're seeing are just probably lagging behind. So the optimistic view is the future is going to be full of exciting new ideas. So Andrew Young is just kind of the beginning of the iceberg and I hope that iceberg doesn't, it's not this influencer. One of the things that really bothers me, I've gotten a chance, not should be careful here, I don't wanna I love everybody, but you know, these people who talk about like, you know how to make your first million or how to succeed and and they're so, I mean, yeah, that that makes me a little bit cynical about uh I'm worried that the people that win the game of Politics will be ones that want to when the game of politics and like we mentioned a L. C. Is I hope they optimize for the 80% populist thing, right? Like they optimize for that badass thing that history will remember you as the great man or woman that did this thing versus how do I maximize engagement today? And keep growing those numbers? The influencers are so, I'm so allergic to this man. They keep saying how many followers they have and the different accounts and it's like, I don't think they understand, maybe I don't understand, I don't really care. I think it has destructive psychological effects. One, like thinking about the number like getting excited. Your number went from 100 to 101 and being like, and today went out to 105, whoa, That's a big jump that maybe like thinking in this way, like I wonder what I did, I'll do that again in this way. One, it uh it creates anxiety, almost psychological effects. Whatever the more important thing is, it prevents you from truly thinking boldly in the long arc of history and creatively thinking outside the box, doing huge actions. And I actually, my optimism is in the sense that that kind of action will beat out all the influencers. [/INST] Well, I don't know, lex this is my cynicism comes in. So there's a guy Madison cawthorn, the youngest member of congress. Um, and he, I don't want to say got caught, but there was like an email where he was like, my staff is only oriented around coms. Like he was basically saying he got basically constant, like my staff is only centered on communications and that's the right play. If you do want to get the benefits of our current electoral political and engagement system, which is that what's the best way to be known within the right as a, as a right wing politician. It's to be a culture warrior. Go on Ben Shapiro's podcast, be one of the people on Fox News, go on sean Hannity's show, go on tucker's show and all of that because you become a mini celebrity within that world Left unsaid, is that, that world is increasingly shrinking portion of the American population and they barely, they can't even win the popular vote election, um, let alone barely win. Eke out an electoral college victory in 2016. Well, but the incentives are all aligned within that and it's the same thing really on the left, but you're right, which is that ultimately, this is, this is why geniuses are geniuses because they buck the short term incentives, they focus on the long term, they bet big and they usually fail. But then when they get big, they, they succeeds spectacularly. The people I know who have done this the best are like a lot of the crypto folks that I've spoken to, like some of the stuff they say, I'm like, I don't know if that's gonna happen, but look, they're like billionaires, right? You're like, so they were right. So it's the way I've heard it expressed is you can be wrong a lot, but when you're right, you get right big. And I mean, I've seen this thing on my career, I mean he took spectacular risk, like spectacular risk and just double down, double down, double down, double down, double down and you can kind of tell to him, I mean, you know better than I do, but like from my observation, I don't think the money matters as right. I just like when I see him, I'm like, I don't, it's nobody works as hard as you do and builds a way that you build, if it's just about the money, it's just, it just doesn't happen. Like nobody wills SpaceX into existence just for the money. It's not worth it, frankly, right? Like he probably destroyed years of his life and like mental sanity, </s>

<s>[INST] you still able to find the guts to take big risks? [/INST] Yeah. See I'm trying to win. Like I think that gives me an advantage frankly where I might make decisions that are financially suboptimal because I'm trying to win. But there's also an inherent advantage to that. Like that. Again, something I watched and learned from a guy like Michael Adamo where he takes advantage of these people playing so passively in these spots where he's like, I don't, I'm not trying to come, I'm gonna win, I'm just gonna bully bulldoze you because I'm not worried about, you know, the small financial mistake of, you know, pay jump. </s><s>[INST] What advice could you give to beginning poker [/INST] players actually </s><s>[INST] at every level how to get better, how to improve, how to improve their game. Obviously, as you said, it's easiest to get better in the beginning, but what advice would you give, how to get better. [/INST] So one of the ways, I mean, I think way back to how I started, right? And there's so many resources and tools available right now to analyze hands, but when you play right, and you find yourself in a situation or a hand that you're not really sure about not because you had aces and went all in and you lost, like that's not interesting, but an interesting situation where you're not sure what you did. Drop, drop the hand down, write it out and then either a, you know, use some of the tools whether it's the solvers if your, you know advanced enough or ask your group, you know, like have a couple of friends at your level and talk through the different decisions and start to learn that way, right? Because those mistakes that you make or those tough, tough hands, that's where the real learning comes from. Like, so that so basically if you're, because you're gonna be in similar scenarios in poker, you're rarely gonna have the identical situation, but you have situations that are </s><s>[INST] similar, you [/INST] know, you raise with ace king, someone three bet, another guy goes all in. Okay, well what do I do in that spot? You know, it's you're gonna have similar situations in the future as well. So figuring that out the more you can do that you chop away at um, you know, different strategical mistakes, you know, you you used to make that you no longer make, </s><s>[INST] are there resources? Like your master classes really is great. Are there books? [/INST] So there was a guy named Michael esposito, This is again for a little bit more advanced players, but it's a book called modern poker theory, I think it's called um which sort of explains game theory right to the novice. Right? So it's it's a little bit, I think if you're new to poker, it's probably above the rim for you, but once you start to get a little better and you won't understand how to do it. It's probably a good resource for as far as books and there's also like tons of people who stream poker, professional players and then you can get in there and you get in on the chat and you start talking, you ask them, you see people, you know, explaining their thought process and things like that, There's so many free resources and of course my master class I think does a good job of sort of compartmentalizing like, you know, how to attack it on a deeper level and you know, we get it. I try to get into what's funny when I did the master class, I asked him, I was like, well, you know, how high end do you want this in terms of poker? And they're like, we want really, really high end. And I was like, are you sure? And then I started to explain really, really high and I'm like, okay, well maybe the one below that, Right? So I try to explain really complex um you know, theory in a more palatable way in english if you will because some of these kids, you hear them talk and you be like, huh, </s><s>[INST] But you also, which is really nice. Give example, hands that really illustrate a point, which is really nice. Um you also wrote a book, I think 10 years ago, power holding strategy, it's interesting to think how much of the stuff in that book still applies, how much doesn't [/INST] listen. I still think the book holds up to a certain, obviously like, you know, it isn't optimal because there's like more advanced strategies? And if you play that way, people will figure out a way to exploit you. But if you're like an average player playing an average buy ins, like that's sort of what I coined like small ball approach absolutely will work. You know, at the highest level, you have to add much more, a lot more bluffing. But overall, I think it's still, you know, for the most part, there's a lot of really, especially with tournaments, there's a lot of really good principles in the book. What's </s><s>[INST] the difference in the dynamics if you could just comment on between heads up poker and when multiple people are in one hand. What an interesting aspect to everything we've been talking about, from uh, game theory, to exploitative strategies, all that kind of stuff. [/INST] So the biggest difference when you play, let's say, nine handed, you know, against eight other players and, you know, heads up is first of all just the type of hands and the number of hands you're gonna have to play. So the way that it works is that there's nine people, two out of the nine hands you have to put in money and the other seven, you can just fold for nothing. Okay, when your head's up, your forced to put money in every single hand. Okay, And there's only one other hand in front of you, which means the ranges of hands that you play is way wider, right? So if you're nine handed right and you're in first position, you're like, all right, what do I need to play? Like a good pair? You know, two high cards suited a big ace, you know, stuff like that. That's it, right? That's what you're gonna play, right? And you're gonna fold all the rest when your head's up. You look at a king and a two and you're like, well, I gotta play this. You know, you're gonna, you're gonna, you're forced to play a lot more hands and a lot more complex situations when you're playing um, heads up because you're gonna be playing much far weaker hands. Queen five jack three, all these types of hands and you're gonna see flops where you're, you're not gonna have the luxury of being like, I'm in there with a premium hand, Queens, Kings, aces. Those are easier to play, right? Very, very strong holdings heads up. You're forced to dance and fight a lot more. You know, you can't sit in the weeds and wait. Hmm, heads up is very intense. I like heads up. But I think if you had to play heads up 8, 10 hours, it's so mentally draining because your face is so many constant decisions. Each and every spot, like you play nine handed, you look at a nine and three, you throw it away hang out for a bit relaxed, a little break and playing heads up, you're like, it's like, boom, you're, it's like, you're in the ring, you know, you're in the octagon and you're facing like Haymakers nonstop </s>

<s>[INST] So you've said the words Nobel Prize over 240 times. Yeah, I mean, it is so strange as a symbol that kind of unites this whole scientific journey, right? I got, it's both sad and beautiful that a little prize, like a little award, a medal, a little plaque, they'll be most likely forgotten by history completely. So some silly list. It's somehow a catalyst for greatness. It resulted in you doing your life's work. [/INST] The dream of it. Would I have done it without the Nobel Prize? I can't necessarily counterfactually state that that would have happened. So no, it definitely has a place. And for me, it is valuable to think about it. But the level of obsession that academics have about it is really, I think it is almost unbalanced becoming unhealthy. And again, I have no, I make no truck with the winners of the Nobel Prize, obviously. Now I've had 11 on the show. And to think about like the one rule. So by the way, right after the denouement of the story, which I'll get to in a bit, how our dreams went down to dust and ashes, I was asked by the Royal Swedish Academy of Sciences to nominate the winners of the 2015 Nobel Prize in physics. So like the one that I theoretically could have been eligible to win in 2016, actually, they asked me to nominate. Now imagine if I ask you, Lex, you say Brian, instead of me inviting myself on the show, if you say, Brian, would you like to come on the Lex Friedman podcast? I say, you know what, Lex, you know that guy, Rogan, I think you might already, can you introduce him to me? Like, do you imagine how that would feel? Like you'd be like, ah, you know, I'm here. So I was asked to nominate the winners. And the one rule that they say, of all the rules that Alfred Nobel stipulated, there's only one rule that they maintained. In other words, he said one person can win it for something they discovered in the preceding year that had the greatest benefit to mankind, made the world better, right? None of that was mentioned in the letter. It said many people can win it for work done long ago. They didn't mention anything in the letter to me, signed by the secretary general, nothing about benefiting mankind. He said, just one thing, can't nominate yourself. </s><s>[INST] So none of these guys nominated themselves. Actually, little known fact, they sent that exact letter just to you. [/INST] That rule was created just for you. That's called the Keating correlate. </s><s>[INST] Yes, exactly. Just to like, good for them, rub it in. I mean, it's in this particular case, of course, there's like some weird technicality or whatever. [/INST] But in this particular case, that's kind of a powerful. </s><s>[INST] Good for them. A reminder that the Nobel prize leaves a lot of people [/INST] behind in their stories behind all of that. Yeah. No way. Yeah, I mean, here's a good example. Again, this is my friend, Barry Barish. He's become like a mentor and a friend. He wrote the forward to this, my book into the impossible. He won the Nobel prize because a different guy died and he admits it and he said it. And actually it's funny with him because I've heard you talk, you know, very rhapsodically and lovingly and romantically about, with Harry Cliff and a wonderful podcast with him, by the way, about the LHC and how wonderful it is and how in that, you know, we were about to build the superconducting supercollider right here in Texas and it didn't get built and got canceled by Congress. And I'd say to barry, that was the best thing that ever happened to you. He's like, what the hell are you talking about? And I'm like, if that didn't get canceled, first of all, the, even though it did get canceled, the Europeans went on to build it themselves, saved the American taxpayers billions of dollars. And we wouldn't have learned anything really substantially new, as proven by the fact that, as you and Harry talked about, nothing besides the Higgs particle of great note has come out. And actually, he's had a recent paper, but it's been an upper limit, along with his collaborators on the LHCB experiment that I'm going to be talking with him about. But the bottom line is, it was really built to detect the Higgs. So the SSC, for twice as much money, would have sucked up Barry's career. He would have been working on that, maybe not. And then he would never have worked on LIGO. And then he wouldn't have won the Nobel Prize, right? So you look at counterfactual history. That's not actually a big stretch, right? If the SSC had still gone on, he would have worked that, because he was one of the primary leaders of that experiment. Second thing, if, imagine the following thing had happened. They won the Nobel Prize, because in September 2015, they detected unequivocal evidence for the in-spiral collision of two massive black holes, each about 30 times the mass of the sun, leaving behind an object that had just less than 60 solar masses behind. So one solar mass worth of matter got converted to pure gravitational energy. No light was seen by them. This particular date, September 14, 2015. That explosion, because of the miracle of time travel that telescopes afford us, that actually took place 1.2 billion years ago in a galaxy far, far away. They actually don't know which galaxy took place in stealth, and they never will. OK? If that collision between these two things, which have probably been orbiting each other for maybe a million years or more, if that had occurred 15 days earlier, </s>

<s>[INST] Would you say that? So youtube, when I upload videos often generates automatic captions, are you sort of from a company perspective from uh tech perspective, are you trying to beat youtube? Google? Hell of a google. I mean I don't know how seriously they take this task but I imagine it's quite serious and they, you know, google is probably up there in terms of their teams on on a. S. Are just NLP natural language processing, different technologies. So do you think you can be google [/INST] on this kind of stuff? Yeah, we think so. Google just woke up on my, this is hilarious. Okay, now google is listening, sending it back to headquarters. Who are these rough people but that's the goal? Yeah, I mean we measure ourselves against like google amazon Microsoft, you know, some of the some smaller competitors um Mhm And we use like our our internal tests with that. We try to compose it of a Pretty Representative set of audience, maybe it's some podcast, some videos, some some interviews, some lecturers, things like that, right? And we beat them in our own testing </s><s>[INST] and uh actually offers automated, like you can actually just do the automated captioning. So like I guess it's like way cheaper, whatever it is, whatever the rates are. [/INST] Yeah, it's </s><s>[INST] by the way, it used to be a dollar per minute for captioning and transcription, Things like a dollar 15 or something like that. $25.25 25 now. Yeah, it's pretty cool. That was the other thing that was surprising to me, it was actually, yeah, like the cheapest thing you could want to, I mean, I don't remember it being cheaper, you could on upward get cheaper, but it was clear to me that this that's going to be really shitty, so like you're also competing on price. I think there were services that you can get, like similar to rev kind of um feel to it, but it wasn't as automated, like the dragon drop the entirety of the interface. It's like the thing we're talking about, I'm such a huge fan of like frictionless, like uh amazon's single By Button, whatever that one click, one click, that's genius right there. Like that is so important for services, that simplicity and I mean revenue is almost there. I mean there's like some trying to think so I I think I've uh I stopped using uh this pipeline, but rather offers it and that, I like it but it was causing me some issues on my side which is um you can connect it to like Dropbox and it generates the files and dropbox so like it closes the loop to where I don't have to go to rev at all and I can download it. Uh Sorry, I don't have to go to rev it all and to download the files, it could just like automatically copy them. [/INST] You're putting a drop like some, you know, a day later or maybe a few hours later shows up just shows up. </s><s>[INST] Yeah, I was trying to do it programmatically too. Is there an api interface you can I was trying to do like through python to download stuff automatically, but then I realized this is the programmer at me. Like dude, you don't need to automate everything like in life like flawlessly because I wasn't doing enough captions to justify to myself the time investment into automating everything perfectly. [/INST] Yeah, I would say if you're doing so many interviews that your biggest roadblock is uh looking on the rough download but now and now you're talking about Elon musk levels of business, but for sure we have like a variety of ways to make it easy, you know, there's the the integration you mentioned, I think the story company called shape here, which kind of can connect Dropbox to rev and vice versa. We have an A. P I if you want to really like customize it, you know if you want to create the lex Friedman, you know uh CMS or or whatever </s><s>[INST] for this whole thing. Okay cool. So can you speak to the the S. Are a little bit more like what is it uh What does it take? Like approach wise machine learning? Why is how hard is this problem? How do you get to the 3% air rate? A good your vision of all of this? [/INST] Yeah well the 3% rate a rate is definitely that's that's the grand vision. Uh we'll see what takes together um But we believe you know in in s are the biggest thing is the data right? Like that's true of like a lot of machine learning problems today right? The more data you have in high quality the data, the better label the data. Um Yeah that does he got good results and we drive have kind of like the best data like we have like you're literally your </s><s>[INST] your literal model is annotating [/INST] the data. Our business model is being paid to annotate that data data data. Uh So it's kind of like a pretty magical flywheel. Uh And so we kind of like reading this flywheel to to this point. Um And we think we're still kind of in the early stages of figuring out all the parts of the fly will to use you know because we have the final transcripts. Um and we have the uh the audience and we train on that. But we in principle also have all the edits that the robbers make right. Um </s><s>[INST] Oh, that's interesting. I can use that as a [/INST] well that's that's something for us to figure out in the future. But you know, we feel like we're only in the early stages so the </s>

<s>[INST] Yeah, they don't want to manipulate each other for sure. Uh that's what I mean. You kind of spoke to ethics the one of the concerns that people have in this modern world, the digital data is that of privacy and security. But privacy, you know, they're concerned that when they share data it's the same thing we feel when we're trust other human beings uh in being fragile and revealing something that we're vulnerable, vulnerable about. There's a there's a leap of faith. There is a leap of trust that that's going to be just between us. There's a privacy to it. And then the challenge is when you're in the digital space and sharing your data with companies that use that data for advertisement, all those kinds of things. There's a hesitancy to share that much data to share a lot of deep personal data and if you look at brain data that feels a whole lot like it's richly deeply personal data. So how do you think about privacy with this kind of ocean of data. [/INST] I think we got off to a wrong start with the internet where the basic rules of play for the for the company that B was if your company you can go out and get as much information on a person as you can find without their approval and you can also do things to induce them to give you as much information and you don't need to tell them what you're doing with it. You can do anything on the back side, you can make money on it but the game is who can acquire the most information and devised the most clever schemes to do it. That was a bad starting place. And so we are in this period where we need to directly to correct for that and we need to say first of all, the individual always has control over their data. It's not a free for all. It's not like a game of hungry hippo but they can just go out and grab as much as they want. So for example when your brain data was recorded today, the first thing we did in the kernel app was you have control over your data. And so it's individual consent. It's individual control. And then you can build up on top of that but it has to be based upon some clear rules of play. If everyone knows what's being collected they know what's being done with it and the person has control over it. </s><s>[INST] The transparency and control. So everybody knows what does control look like. My ability to delete the data [/INST] if you want to delete it and to know who is being shared with under what you know what under what terms and conditions we haven't reached that level of sophistication with our products of if if you say for example hey Spotify please give me a customized playlist according to mine. you're um you know, you could say you can have access to this vector space model but only for this duration of time and uh and then you've got to delete it. We haven't gotten there at that level of sophistication, but these are ideas we need to start talking about of how do you, how would you actually structure permissions? And I think it creates a much more stable set for society to build where we understand the rules of play and people aren't vulnerable to being taken event. It's not fair for an individual to be taken advantage of without their awareness with some other practice that some companies doing for their sole benefit. And so hopefully we are going through a process now we're correcting for these things and that it can be an economy wide shift that because really these are these are fundamentals we need to have in place. </s><s>[INST] It's kind of fun to think about like in chrome when you install an extension or like install an app, it's ask you like what permissions you're willing to give and be cool if in the future it is like uh you can have access to my brain data. I [/INST] mean it's not unimaginable in the future. The big technology companies have built a business based upon acquiring data about you that they can then create a model of you and sell that predictability. And so it's not unimaginable that you will create with Colonel device for example, a more reliable predictor of you than they could and that they're asking you for permission to complete their objectives and you're the one that gets to negotiate that with them and say sure, but it's not unimaginable, that might be the case. </s><s>[INST] So there's a guy named Elon musk and he has a company in one of the many companies called neuralink that has uh that's also excited about the brain. So be interesting to hear your kind of opinions about a very different approach that's invasive that requires surgery implants a data collection device in the brain. How do you think about the difference in kernel and your link in the approaches of uh getting that stream of brain data. [/INST] Ellen and I spoke about this a lot early on we we met up, I had started Colonel and he had an interest in pre interfaces as well and we explored doing something together, him joining Colonel. And ultimately it wasn't the right move. And so he started, not only can I continue building Colonel, but it was interesting because we were both at this very early time where it wasn't certain what if there was a path to pursue if now was the right time to do something and then the technological choice of doing that. And so we were both our starting point was looking at invasive technologies and I was building to invasive technology at the time that's ultimately where he's gone ah a little less than a year after uh you and I were engaged, I shifted Colonel to do non invasive And we had this neuroscientists come to colonel, we were talking about he had been doing a neurosurgeon for 30 years, one of the most respected neuroscientists in the us. And we brought him to colonel to figure out the ins and outs of his profession. And at the very end of our three hour conversation, he said, you know, Every 15 or so years a new technology comes along. That changes everything. He said. It's probably already here. You just can't see it yet. And my jaw dropped. I thought because I I had spoken to bob Greenberg who had built uh second sight first on the optical nerve and then he did court up an array on the optical um cortex. And then I also became friendly with um no pace. Who does, who does the implants, first, seizure detection and remediation. And I saw in their eyes, What it was like to take something through an implantable device through for her for a 15 year run. They initially thought was seven years, Ended up being 15 years and they thought would be 100 million is, you know, 300, million. And I really didn't want to build invasive technology. It was the only thing that appeared to be possible. But then once I spun up an internal effort to start looking at non invasive options, we said, is there something here? Is there anything here that, again, has the characteristics of it has a high quality data, it could be low cost that could be accessible, couldn't make brain or faces mainstream. And so I did a bet the company move, we shifted from non invasive to invasive to non invasive. </s>

<s>[INST] old were you to hear? So 7th grade, 8th Grade 8, 15. That is, that's very good. [/INST] Yeah. And it's, uh, that's romanticism, you know, with a little bit of like too much. It's musically cheesiness. A little bit, I would say, but very strong. </s><s>[INST] So let's go, let's go to [/INST] Kiev. Okay, let's go to Kiev. </s><s>[INST] You were born and raised in Kiev. Just like you said with the guys with the rocks and the rockets. [/INST] How about that? It was stupid guys. You know, I got already got </s><s>[INST] that. You can hold a grudge. What are some memories of your dad and mom that stand out and [/INST] mm. Okay. Yeah. What we do? Well let's drink japanese. </s><s>[INST] Yeah. What are we drinking? Japanese whiskey, [/INST] japanese whiskey. Uh they call it from the barrel. </s><s>[INST] We're not sponsored by nickel whiskey. Good whiskey. [/INST] Yeah, we do not. But they're doing very, very good job and they are in sub portal. It's a Al Qaeda north of Japan. So they are doing good beer which is Sapporo, very good, very good whiskey and a lot of red caviar can make sure that's it. </s><s>[INST] Did you ever think when you were uh that shortage of bread and you're doing the rocket that you would be sitting here in America a few years later drinking japanese whiskey. Talking about caviar, did you even know what caviar was? [/INST] No, no, obviously because failure it was for holidays a little bit and it was distributed. It's not like you go and buy it. Mhm but you got it a little bit like a, a small can. So we all of us, we knew that and just, it was very popular and </s><s>[INST] well let me just actually ask that question. How did you imagine what, Where you would be 50 years from then? Like how did you think about your future? [/INST] Actually like uh in about eight grades nine grades. I was very very, very, very, very, very good in physics mathematics and chemistry. So I was actually in this system of like Olympia, so called competition. I was absolutely number one in Ukraine. Twice An absolute number two in the soviet union also, twice </s><s>[INST] this is physics competition or mathematics. It was physics for people who don't know, maybe you can explain. I mean this is still goes on to this day. These are very intense competitions. I don't know if they're that popular in the United States. Probably because the United States folks don't do is great but there [/INST] it is extremely popular but not in physics. What is going on in this country? I was very much impressed. It's very serious and very tough competition and mathematics in math, that's right. Uh in physics last. Biology last </s><s>[INST] So you already know knew someone in eighth grade that you're very good at. You said wait a minute, you said chemistry to physics Chemistry and you competed you want in Ukraine [/INST] and I was number two in soviet Union was twice now very interesting that in soviet Union was a rule you're supposed to choose. So you can participate in this competition and Physics, Mathematics and Chemistry and all three of them up to the level of Championship of the region of the big city. In my case it was Kiev. But then to go to competition in the Ukraine You have to choose, you can go to only 2, 1. Mm So I want physics and Chemistry and mathematics in Kiev, then I choose physics and I was going to Ukrainian. Now if you win in Ukraine, so a couple of people from the whole Ukraine goes to this championship of soviet union. </s><s>[INST] That's an incredible accomplishment by the incredible it was, I mean, you basically, you peaked in terms of physics and poetry, age [/INST] 14, that's a big difference, you know, in physics, I was really on top of the Soviet Union, which is a huge country in poetry. It was just no, I'm not a professional, you know, and </s><s>[INST] you can't compete in poetry. [/INST] Yeah, well, basically over there, gold medal from high school, what do you call here? Valedictorian? </s><s>[INST] Which is hard to do. I [/INST] would say here, it's also very not so easy independent in school, but in Kiev, it was an interesting situation because you have to the most difficult was actually not physics, mathematics and it was writing because it was writing not on the computer without spell check, just and you have to write Uh one. Let me see, let me see. Yeah, 12, 12 ah pages as a minimum of an essay. And it's supposed to be zero mistakes. And the Russian language is not the easiest one. </s><s>[INST] And when you say mistakes, I mean, this is what makes a mistake. It's not just, it's grammar, it's writing mistakes [/INST] to writing mistake, uh incorrect use of </s><s>[INST] word or something. Can you cross stuff out? No, right, This is the you can't make a mistake in the actual writing process, you're writing with a pen. Uh [/INST] and first of you were not supposed to make any, </s><s>[INST] you're not allowed to cross the lines and it's perfection in in all the [/INST] silliest for them. And </s><s>[INST] it's silly, but maybe what do you think about that? What what do you think about that strictness? Because it, it, it's easy to criticize from a distance, but there's something powerful about it. [/INST] You know what countries more powerful from this point of view is </s><s>[INST] japan. It's [/INST] japan strictness. I was so much impressed with their kids. What they keep saying that in Japan for kids, it does matter, it's a holiday or the weekend or whatever each day you're supposed to train with these japanese characters because if not it's just impossible just to, I remember it. So there they're working too hard, but as a result they can sell for them quality because quality means no mistakes, right? It's very interesting point, what does mean quality? It means enormous, isn't it perfectionism and this perfectionism creates this kind of whiskey. Yeah. </s>

<s>[INST] on this from a different angle with the cellular automata stuff. It's so suspicious how well it [/INST] works. Yeah, but on their hand, it's like, yet we're still also in a prime number maze. There's </s><s>[INST] things we just don't understand. Within the constraints of the non-prime numbers, [/INST] we find math to be extremely effective, surprisingly effective. Yeah, exactly. Maybe the math we have gets us through the equivalent of the even turns and the odd turns, but there's math we don't yet have that is more complex or more complex rules for other parts. Displaying all our rage, we're all just rats in a cage. I know that gets very abstract, but there are unsolved problems in physics, like the condensed matter space, there's a lot of interesting stuff happening. My recollection, I may be out of date on this, things like sonoluminescence, we don't know exactly how they work. Sometimes those things that are at the edges of physics, in the late 1800s, somebody famous, I think it was Rutherford said, basically all physics is being discovered, et cetera. That was obviously before quantum mechanics. That sort of edge case, people are looking at the bomber and the passion series and seeing this weird thing with the hydrogen spectrum and it was quantized and that led to the sort of phenomenological set of observations that led to quantum mechanics and everything. Sometimes I think the UAP stuff might be like that. People immediately go to aliens for UAP, like the unidentified aerial phenomena. There's surprising amount of stuff out there on this. The UK has declassified a bunch of material. Harry Reed is the sender has talked about this. It's not an obviously political thing, which is good. It's something that, is there something happening there? People have thought for a long time that the UAP thing was American counter propaganda to cover up their new spy planes that were spying over </s><s>[INST] the Soviet Union to make anybody who talked about them seem- Despite all our rage, [/INST] we're all just rats in a cage. Crazy and hysterical or whatever, but if the UAP thing is real, it could be atmospheric phenomena like the Aurora Borealis or the Northern Lights, but some things we don't understand. It could be something like the bomber and passion series, which were the observations of like emission spectra before quantum mechanics. So that's another option as opposed to doesn't exist or little green men. It could be physics </s><s>[INST] we don't understand yet as one possible. Do you think there's alien civilizations out there? [/INST] So there's a lot of folks who have kind of written and talked about this as the Drake equation, which is like multiplying all the probabilities together. There's perhaps more sophisticated takes like the dark forest, which says that if the universe is like a dark forest, we're the dumb ones that aren't hiding our presence. There's one calculation I saw, and I haven't reproduced it myself, but basically says that the assumption that other civilizations have seen ours is wrong because when you have a spherical radius for the electromagnetic radiation that's leaving our planet, as that sphere gets larger and larger, it gets smaller and smaller amounts of energy. So you get farther out, you're not getting enough photons or what have you to actually detect it. I actually haven't looked into the math behind it, but I remember seeing that argument. So actually it is possible that it's so diffuse when you go past a certain number of light years out that people, that an alien civilization wouldn't </s><s>[INST] be able to detect it. That's another argument. That's more basically about signals from them, from us, to be able to signals colliding enough to find the signal from the noise. Right, exactly. Intelligence signal. [/INST] Yeah, Hansen has an article called Grabby Aliens. Have you seen his thing? He's been on this podcast. Oh, great. He's brilliant. I like him. He pushes boundaries in interesting ways. In every ways, in all of the ways. In all the ways, that's right. I like him overall. He's an asset to humanity. </s><s>[INST] He's been on this podcast. Grabby Aliens, so he has this interesting idea that the civilizations quickly learn how to travel close to the speed of light, so we're not going to see them until they're here. [/INST] Yeah, that's possible. Here's, for example, a mystery that we haven't yet done, we haven't really figured it out yet, which is abiogenesis in the lab. We've done lots of things where you can show macromolecules binding to each other, you can show evidence for the so-called RNA world, abiogenesis to go from non-life to life in the lab. You can show microevolution. Obviously, with bacteria, you can do artificial selection on them. Lots of other aspects of fundamental biochemistry origins of life stuff have been established. There's a lot of plausibility arguments about the primitive environment and nitrogens and carbons snapping together to get the RNA world is the initial hypothesis. But to my knowledge at least, we haven't actually seen abiogenesis demonstrated. Now, one argument is you need just this massive world with so many different reps before that actually happens. One possibility is if we could do atomic level simulations of molecules bouncing against each other, it's possible that in some simulation we could find a path, a reproducible path to abiogenesis and then just replicate that in the lab. I don't know. But that seems to me to be a mystery that we still don't fully understand, like an example of the prime number maze. One of the most fascinating mysteries. One of the most important. And again, there may be some biochemist who's like, oh, I apologize, you didn't know about X, Y, and Z that happened in the abiogenesis field. I freely confess I'm not like, oh, Courant on it. The last thing I remember looking at is- What does Courant mean? Like up to the moment, right? </s>

<s>[INST] and the brain has a similar kind of mechanism for doing so that's kind of interesting and that how's that all encoded in the brain, like where is the storage of information, are we talking? Just maybe to get it? Uh a little bit more specific. Is it in the hardware of the actual connections. Is it in chemical communication is an electrical communication [/INST] to do. So this is you know a paper that we are bringing out soon which was um this is the cortical microcircuits paper that I sent you a draft off. Of course this is a lot of it is still hypothesis. One hypothesis that you can think of a cortical column as encoding a concept concept. You know, think of it as a uh an example of a concept is um is an edge person or not or is is an object person or not? Okay so you can you can think of it as a binary variable, binary random variable. The presence of an edge or not or the presence of an object or not. So each cortical column can be thought of as representing that one concept one variable. And then the connections between these cortical columns are basically encoding the relationship between these random variables. And then there are connections within the cortical column. There are each cortical column is implemented using multiple layers of neurons with very very very rich structure there you know there are thousands of neurons in a cortical column. </s><s>[INST] But but that structure is similar across the different cortical column, [/INST] correct, correct. And also these cortical columns connect to a sub structure called thalamus in the uh so all all cortical columns passed through this sub structure. So our hypothesis is that yeah the connections between the cortical columns implement this uh you know that's where the knowledge he stood about, you know how these different concepts concepts connect to each other and then the neurons inside this cortical column and in thalamus in combination implement these uh actual computations in there before inference which includes explaining away and competing between the different hypothesis. Um And it is already. So what is amazing is that uh neuroscientists have actually done experiments to the tune of showing these things. They might not be putting it in the overall inference framework, but they will show things like if I poke this higher level neuron uh it will inhibit through this complicated look through the thalamus, it will inhibit this other column. Uh So they will do </s><s>[INST] such experiments. But do they use terminology of concepts for example? So so your your I mean uh is it uh is it something where it's easy to anthropomorphize and think about concepts like you start moving into logic based kind of reasoning systems? So um are we to think of concepts in that kind of way or is it is it a lot messier, A lot more gray area? You know, even even more gray, Even more messy than the artificial neural network kinds [/INST] of abstractions, easiest way to think of it as a variable, right? It's a binary variable which is showing the presence or absence off </s><s>[INST] something something. But I guess what I'm asking is is that something uh we're supposed to think of something that's human interpret herbal of that something [/INST] it doesn't need to be, it doesn't need to be human interpret herbal, there's no need for it to be human interpret herbal. Uh but it's it's almost like um you you will be able to find some interpretation of it because it is connected to the other things, </s><s>[INST] you know? And the point is as useful somehow, it's useful as an entity in the graphic, in connecting to the other entities that are let's call them concepts. Okay, so by the way, what are these the cortical microcircuits? [/INST] These are the cortical microcircuits, you know, that's what neuroscientists used to talk about. The circuits in in within a level of the cortex. So you can think of, you know, let's think of in neural network, you know, artificial neural network terms, you know, people talk about the architecture of the, you know, so how many, how many layers they build, you know, what is the fan and find out etcetera. That is the macro architecture. So, and then within a layer of the neural network you can, you know, the cortical neural network is much more structured with within a level, there's a lot more intricate structure there. But even um even within an artificial neural network you can think of in feature detection plus pulling as +11 level. And so that is kind of a micro circuit. Uh it's much more complex in the real brain. Uh so within a level, whatever is that circuitry within a column of the cortex and between the layers of the cortex. That's the micro circuitry. </s><s>[INST] I love that terminology, machine learning people don't use the circuit terminology but they should. It's nice. So Okay. Uh Okay, so that's uh that's the cortical microcircuits. So what's interesting about what can we say? What is the paper that you're working on? A proposal about the ideas around these cortical microcircuits? [/INST] So this is a fully functional model for the micro circuits of the visual cortex. </s><s>[INST] So the paper focuses on your idea and our discussion now is focusing on vision. Yeah, the visual cortex. Okay, yeah, this is a model. This is a four model. This is this is how [/INST] vision works. But this is this is a hypothesis. Yeah. Okay, so let me let me step back a bit. Um so we looked at neuroscience for insights on how to build a vision model and and and we synthesized all those insights into a computational model. This is called recursive cortical network model that we used for breaking captures and uh and we are using the same model for robotic picking and uh tracking of objects. And </s>

<s>[INST] coaching him as an MMA fighter. So almost like the psychological aspect [/INST] of doing the best you can for him. </s><s>[INST] Exactly, yep, yep, yep. What was the hardest about the actual training? Was it the technical aspect of trying to figure out the puzzle of Islam, or was it like being a good training partner in figuring out how the grappling would work, basically playing your best impression of Makachev? Were you trying to actually impersonate him? No, not just visually, but like in style. Yeah, definitely, definitely visually. You're not as good looking, but go ahead. [/INST] Yeah, a little taller, yeah, a little taller, but in terms of the training, yeah, I mean, Islam's known as incredibly strong glass. Obviously, I'm heavier than Islam, so theoretically, I should be able to replicate that strength difference. And then in terms of grappling, targeting those submissions that Islam does, like focusing on those in the training room, focusing on the way he holds half guard, and really in the grappling sense, trying to replicate him on the ground, and then, yeah, I wrestle with him on the wall a ton, trying to replicate, obviously, to the best of my ability, a lot of the stuff he does on the wall, body lock heavy, inside trip, Uchimadas, and just constantly putting the work on vox, you know what I mean? Like constantly chaining attacks against him, </s><s>[INST] really replicating that. As he's trying to get up and escape and all that kind of stuff, so all the submissions, [/INST] both judo and submissions, just attack and attacking. Exactly, and there's only so much you can do really, because obviously he's been, I think he's been fighting a long time, so it's like you're trying to polish what he already is good at. You can't just completely create an entirely new game for him in the space of six weeks. So you're trying to take what he's already effective at, add to it. And luckily, a lot of the stuff he's already very good at </s><s>[INST] was easy to add to for the fight. Question from Reddit. I'm very curious why other MMA fighters don't employ high-profile grapplers from B Team and New Wave to improve their grappling, this from the subreddit. By now, it's clear that they are levels above almost everyone in MMA, simply because fighters there don't specialize in grappling, but it doesn't seem like fighters, even champions, get training partners from the most successful teams. Why is that the case? From your experience, why doesn't it call you? [/INST] You might now, nah, nah, put in a good word for me. Oh, I will. Yeah, that's right, he takes a joke pretty well. Yeah, yeah, no, no, you'll be walking with open arms. I think your average Jiu-Jitsu coach, MMA fighters have bad experience with Jiu-Jitsu guys. Jiu-Jitsu doesn't have a massive place in MMA. Obviously, rounds, stand-ups, it's hard to submit people. Your average Brazilian Jiu-Jitsu black belt doesn't know anything about holding a guy down, doesn't know anything about how to stand up. So I think if you overly utilize that Jiu-Jitsu guy that hasn't had experience in more modern no-gi or training MMA fighters previously, it's gonna be a complete waste of time to them. </s><s>[INST] I think they're smart enough to realize that. Do you have, and do you guys, do you have interest in MMA at all? Just not even like, well, certainly just competing yourself, [/INST] but just understanding the puzzle of MMA. Yeah, I mean, I've been a fan of mixed martial arts for a very long time before I trained Jits. Personally, I'd much rather coach than fight, but I mean, I'd fight somebody for a good check and I get to pick the opponent, have a proper camp. Okay. I can think of a good Atlanta, you know. What you doin', what you doin'? Who's that? All right. </s><s>[INST] Who do you think is the greatest MMA fighter of all time? Craig, I'm gonna start with you. Just as a fan of the sport, [/INST] it's somebody who's been deep in it. I mean, from my perspective, after that performance, I'm gonna say Volkanovski, because he was able to decisively out-strike Max Holloway, one of the greatest strikers in the sport, and he was able to hang with the wrestling of Islam, Akhachev. And in terms of Ortega, he was able to survive Ortega, who has some of the most dangerous missions in Jiu-Jitsu. </s><s>[INST] So I think, in my opinion, technically, he's the best. So even though he technically lost, [/INST] he still has the crown. He's proven himself. Given the size difference given he's movin' up in weight. Yeah. I think all those facts, </s><s>[INST] I believe so. He's proven himself, really. The underdog, everything, the pressure. [/INST] Did you think he would be able to hang in any of the wrestling exchanges with Islam? </s><s>[INST] No, no, no, no, I was really surprised. That's why in my eyes, like, it's kind of funny, like, winning at the end of the day, I feel like judges influenced that. Although I did think Australian judges would rob the other way, but I was assuming. They kind of, somebody paid somebody and not enough, maybe. But in general, I just thought he won sort of in the eyes of what martial art stands for. Like, sort of go into the fire and survive and thrive and finish the last round strong, which is kind of, like, spiritually is what a victory is. So I wish we kept going, like. One more round. Yeah, exactly, that kind of thing. Well, what about you? What do you think? Like, who are the fighters you admire? [/INST] Like, who do you think is the greatest of all time? Winning. One more round. Yeah. I think the fighter I paid most attention to was John Jones. You know, he has a great ability just to mix the high level striking, high level grappling. Although his, you know, jiu-jitsu by itself isn't, you know, probably isn't like, you know, super high level, but his ability to mix everything together, I would say he's the best. And he's a fellow, you know, heavier guy, heavy weight now. So it's nice to see, you know, how those guys move </s>

<s>[INST] to get there. You're just a genius in this whole space of strength and health and almost like biology, that the strength feet is just one representation of that, but this particular strength required, that kind of singular focus which I think, I don't know there's something beautiful about that singular focus that is often only truly perfected athletics. I see it with the greatest olympic athletes as well. The kind of singular focus required, there is incredible and somehow some of the most beautiful things that humans can do and achieve just [/INST] that thing. So that's the thing. It's like, oh that must be when we say singularity of focus, it's not like here because it covers a vast array of stuff. Like I was working with people are, you know all well, yeah, all around North America, I wouldn't say anybody around the globe, but professionals coming in working on different aspects of rehab and, and recovery and like, I mean I'm tapping all sorts of stuff in so many platforms from nutrition to drugs to again, like, you know, various chinese medicine, you know, as far as you know what the humans </s><s>[INST] and just love and positivity and just inspiration, all those kinds of aspects. I mean you probably would have done much more if you went outside North America talked to some Russians just between you and uh some Russians possibly they could give you some, uh I don't know those, there's some incredible strength athletes in Eastern [/INST] europe. Absolutely, I've got the best one coming in september to get fixed. So </s><s>[INST] what do you mean by fixed? [/INST] So I'm not sure what his particular issues are, but he has held the all time world record repeatedly for a long time and he hasn't competed for some time and he just reached out saying he would like to come and have me take a look and see if I can get him fixed because </s><s>[INST] he used to return. Okay, so it's more injury centric versus like form and fundamental centric combination of everything. [/INST] Everyone always wants to focus on the output. How do you, how do you give me the fix for that? But it ties right back into all those other things. Right. So, but yeah, the eastern, the eastern bloc continued to be a dominant force in regards to athletics and strength athletics. Without a doubt, some of my big rivals in my competitive days were, that's who it was. </s><s>[INST] Rivalry brings out the best in us. Can you tell me the story of your childhood? [/INST] It's definitely outside the scope of the norm? Well today, maybe not 150 or 200 years ago, but uh, my parents highly intelligent, you know, people coming out of the bay area. My mom was, you know, going to school to be a chemical engineer, she was a top top student athlete, graduated at her school. My father was a member of MENSA. My stepfather was just a genius but not able to really function in society, but my mom was, you know, she had some demons and some other stuff and just, she just said one day she's like, I just don't want to be part of society. She still isn't uh lives out in the desert but has her mind, but she wanted to figure out a way to make a life outside of that. And so that's where we ended up is up in the mountains in northern California. And a lot of that was, you know, them trying to get into successfully growing marijuana, which back in that, you know, wasn't legal back then, highly illegal. And in fact those areas where some of the areas where lived were quite dangerous. So there's a documentary, um, Murder mountain that came out recently. If you watch that, you'll tie into my book. Uh, just the understanding of the stuff that I was talking about, dealing with serial killers, human trafficking, police corruption, uh, murderers. Like just how real that stuff is if it doesn't capture you from the book. Okay. </s><s>[INST] The book, by the way is the Eagle and the Dragon. [/INST] Yeah, Thank you. A terrible salesperson. Like I told you. Uh, </s><s>[INST] but a good, uh, it's a good title. I don't know if you came up with it, but I didn't. So yeah, we'll talk about that [/INST] anyway. We're living by a stream, You know, Alpha Meadow. There's no roads into where you have to hike in and we've got beams lashed into the trees up above us because that's where our betting is because there's rattlesnake dens all around and six years old. I'm being taught how to capture and handle live rattlesnakes because that's what I need to do to be safe and you can imagine six years old sitting there with a live rattlesnake in your hand, grabbing it by the side of the head control. And so it can't, can't bite you. And it's just wrapping itself around your arm and you're staring at like it's only intent is right then, is to kill you like that's it. Right? Um, you want to take a bath, It's filling up the jug in the stream and setting it out on the rocks during uh, during the sun. So you dump it over your head and you know, not all the living was that way, you know, good part was similar to that tent living living in a 16 ft trailer with a family of six, which is not much bigger than the space that we're sitting here. Um, so we're talking hard winters with feet of snow on the ground, nowhere to go. Um, I'm living in the back of the pickup truck and uh, just a standard sleeping bag that we get from the Salvation Army, not the, not the below zero. So I'm, uh, I'm not sleeping well. There's living in homes that were maybe uh, condemned. There's no no doors even on them, no electricity or running water or one or the other. Both. And sometimes a little bit better. By the time we got to high school, uh, we had a mobile home. So my stepfather had won a disability payment cause he had a broken arm that whole time from a accident a long time ago and finally got an award and got a down payment on this mobile home that didn't have again, doors on the inside, it did have running water, did have electricity, didn't have a kitchen, you know the windows would crank close and open but they wouldn't close all the way. So the trim them in with plastic to be able to try to protect from the elements that was my environment. Like learning how to forage for mushrooms. I mean there was summers I would send and my parents would be out, they were in the drug trade earlier, we got taken by the by the police and put into foster care for for a while which ties into some of the stories with human trafficking and honestly it's in my book, but it's really hard for me to talk about that stuff. Um And obviously not all that's in the book. So um but they got us back, we moved to Oregon and they stayed out of the drug trade from that time to ensure that they didn't lose us again. But quickly we kind of fell back into the same thing. So at that point it was learning about geology and starting to do mining and firewood cutting, but mostly the mining, because Pat's broken arm chains all made a little tough. </s>

<s>[INST] Yeah, you've opened the door about to a bunch of questions I want to ask. But let's first return to the uh, to your tweet and read it like a Shakespeare. You wrote Gado is not the end, it's the beginning and then he wrote moe and that an emoji of a cat. Uh, so first two questions first, can you explain the mo and the cat emoji and second, can you explain what Gato is and how it works? [/INST] Right. Indeed. I mean, thanks, thanks for reminding me that we're all exposing on twitter and </s><s>[INST] permanently there [/INST] permanently one </s><s>[INST] of the greatest ai researchers of all time mo and cat emoji. There you go. [/INST] Right. So you </s><s>[INST] can imagine like touring tweeting mo and probably he would probably would [/INST] probably. So yeah, the tweet is important actually. You know, I put thought on the tweets, I hope people which part. </s><s>[INST] Okay, so there's three sentences. Gatto is not the end. God is the beginning. Cat emoji, which, which is the important part. [/INST] No, no, definitely. Um, that it is the beginning. I mean, I, I probably was just explaining a bit where the field is going, but let me tell you about Gatto. So first, the name Gatto comes from maybe a sequence of releases that the mine had that named like used animal names to name some of their models that are based on this idea of large sequence model. Initially they're only language but we're expanding to other modalities. So we had we had go for team Killah these were language only. And then more recently we released flamingo which adds vision to the equation and then Gato which adds vision and then also actions in the mix right? Um as we discussed actually actions especially discrete actions like up down left, right. I just told you the actions but their words. So you can kind of see how actions naturally map to sequence modeling of words which these models are very powerful. So Gato was named after I believe I can only from memory. Right? These you know these things always happen with an amazing team of researchers behind So before the release we had a discussion about which animal would we pick? Right? And I think because of the word general agent. Right? And and this is a property quite unique to Gatto. We we kind of were playing with the G. A. Words and then you know Gato. Yes. And so is obviously a spanish version of cat. I had nothing to do with it although I'm from spain </s><s>[INST] wait sorry how do you say cat in [/INST] spanish? Got. Oh </s><s>[INST] yeah. Okay. [/INST] Okay. I see, I see I see. You </s><s>[INST] know it all makes sense. Okay. How do you say? Meow in spanish. Now that's [/INST] I think you you say the same way but you're right. It is M I A U. </s><s>[INST] Universal. Alright. So then how does this thing work? So you said general is so you said uh language, vision [/INST] action, </s><s>[INST] how does this, can you explain what kind of neural networks are involved? What does the training look like? Maybe um what you are some beautiful ideas within the system? [/INST] Yeah so maybe the basics of gateau are not that dissimilar from many, many work that comes. So here is where the sort of the recipe hasn't changed too much. There is a transformer model that's the kind of recurrent neural network that essentially takes a sequence of modalities, observations that could be words, could be vision or could be actions and then its own objective that you train it to do when you train it is to predict what the next anything is and anything means what's the next action. If this sequence that I'm showing you to train is a sequence of actions and observations. Then you're predicting what's the next action and the next observation. Right? So you you think of this really as a sequence of bytes, right so take any sequence of words, a sequence of inter leaf words and images, a sequence of um maybe observations that are images and moves in a target up down left right. And this you just think of them as bytes and you're modeling what's the next bite going to be like and you might interpret that as an action as an action and then play it in a game or you could interpret it as a word and then write it down if you're chatting with the system and so on. Um So Gato basically can be can be thought as inputs images, text, video actions. It also actually inputs some sort of um pro preconception sensors from robotics because robotics is one of the tasks that it's been trained to do. And then at the output similarly it outputs words, actions. It does not output images. That's just by design. We decided not to go that way for now. Um That's also in part why it's the beginning because there's more to do clearly but that's kind of what the Gato is. Is this brain that essentially you give it any sequence of these observations and and modalities and it outputs the next step and then you off you go you feed the next the next step into and predict the next one and so on. Now it is more than a language model because even though you can chat with gatto, like you can chat with or flamingo. Um It also is an agent. Right? So that's why we call it a of Cotto like the word, the letter A and also its general um It's not an agent that's been trained to be good at only Starcraft or only Atari or only go it's been trained on a vast variety of datasets. So </s><s>[INST] what makes an agent if I may interrupt the fact that I can generate actions. [/INST] Yes. So when we call it? I mean it's a it's a good question. Right? What why when do we call a model? I mean everything is a model. But what is an agent in my view is indeed the capacity to take actions in an environment that you then send to it and then the environment might return with a new observation. Um and then you generate the next action. And </s>

<s>[INST] I believe the editors used to have much more [/INST] power and he was a young man whose name was Tate and he ended up being editor for years. But uh so he sent this for review to uh a theoretical physicist named Robertson who was also in this field of general relativity, who happened to be on sabbatical at that moment at caltech. Otherwise his institution was Princeton where I sign was and he saw that the way they set up the problem, the infinities were like I make it as a student because if you don't set it up bright and general relativity you get these infinities. And so he reviewed the article and told gave an illustration that if they set it up in what are called cylindrical coordinates, these infinities went away. He's uh Editor of Physical Review was obviously intimidated by Einstein. He wrote this really not a letter back, like I would get saying, you know, you're screwed up in your paper instead it was kind of uh what do you think of the comments of our referee Einstein wrote back, it's a well documented letter wrote back a letter to physical review saying uh I didn't send you the paper to to send it to one of your so called experts. I sentence you to publish, I know I withdraw the paper uh and he never published again in the in that journal, that was 1936. Instead he rewrote it with the fixes that were made, changed the title and published it in what was called the franklin Review, which is the franklin Institute in philadelphia, which is Benjamin franklin Institute, which doesn't have a journal now, but did at that time. So the article is published, it's the last time he ever wrote about it, it remained controversial. So it wasn't until Close to 1960, where there was a conference which brought that brought together the experts in general relativity to try to sort out whether there was um whether it was true that there were gravitational waves or not. And there was a very nice derivation by a british theorists from the heart of the theory that gets gravitational waves. Uh And that was number one, the second thing that happened at that meeting is Richard Feynman was there and Feinman said, well if there's a typical Feinman, if there's gravitational waves, they need to be able to do something otherwise they don't exist. So they have to be able to transfer energy. So he made a idea of a good Duncan experiment, that is just a bar with a couple of rings on it and then if a gravitational wave goes through it distorts the bar. Uh and that creates friction on these little rings and that's heat and that's energy. So that that meant, </s><s>[INST] is that a good idea? That sounds like a good idea. [/INST] Yeah. It means that he showed that with the distortion of space time, you could transfer energy just by this little idea. And it was shown theoretically. So at that point it was uh huh believe theoretically than by people that gravitational waves should exist. </s><s>[INST] No, we should be able to detect that [/INST] means we should be able to detect them except except that they're very, very small. </s><s>[INST] Just so what kind of uh there's a bunch of questions about what kind of events would generate gravitational waves? [/INST] You have to have. This, what I call quadruple moment that comes about. If I have uh for for example, to objects that go around each other like this, like the Earth, the Earth around the sun, or the moon around the Earth. Or in our case it turns out to be two black holes going around each other like this. </s><s>[INST] How is that different than basic oscillation? Back and forth. This is just more common in nature. [/INST] Oscillation is a dipole moment. </s><s>[INST] It has been three dimensional space kind of [/INST] So you have to have something that's three dimensional, that'll give what I call the quadruple moment that's just built into this </s><s>[INST] and luckily nature you have stuff uh [/INST] and luckily things exist and it is luckily because the effect is so small that you could say look I can take a bar bell and and spin it right and detected gravitational waves. But unfortunately no matter how much I spend it, how fast I spent it it's so I know how to make gravitational waves but they're so weak. I can't detect them. So we have to take something that's stronger than I can make otherwise. We would do what hurts did for electromagnetic waves? Go in our lab, take a bar bell, put it on something spinning </s><s>[INST] casket. Don't question. So a single object that's weirdly shaped. Does that generate gravitational waves? So if it's if it's rotating [/INST] sure it was just much weaker so it it's weaker. Well we didn't know what the strongest signal would be that we would see, we targeted seeing something called nutrients stars actually because black holes we don't know very much about it turned out were a little bit lucky there was a stronger source which was the black holes. </s><s>[INST] Well, another ridiculous question. So you say waves, what does what does the wave mean? Like uh the most ridiculous version of that question is what does it feel like to ride a wave as you get closer to the source or experience it. [/INST] Well if you experience a wave, imagine that this is what happens to you. I don't know what you mean about getting close. It comes to you. So it's like it's like uh this lightweight for something that comes through you. So when the light is you makes your eyes detective. I flashed it. What does this do is it's like going to the amusement park and they have these mirrors, you look in this mirror and you look short and fat and the one next to you makes your tall and thin. Okay. Imagine that you went back and forth between those two mirrors once a second that would be a gravitational wave with a period of once a second. Uh if you did it 60 times a second, go back and forth. And and then that's all that happens. It makes you taller and shorter and fatter back and forth as it goes through you at the frequency of the gravitational wave. So the frequencies that we detect are higher than one a second. But that's the idea. But and this amount is small </s>

<s>[INST] Okay. Okay. I'm not sure that's true, but that statement gives me help. Does this ultimately excite you about the future, or does it terrify you? So let's, we mentioned that time is fundamental. It seems like you're at the cutting edge of progress that will have to happen, that will happen, that there's no stopping it. And I, as we've been talking about, I see obviously a huge number of exciting possibilities. So whenever you automate these kinds of things, just the world opens up. It's like programming itself and the computer, regular computer has created innumerable applications and made the world better in so many dimensions. And it created, of course, a lot of negative things that we for some reason like to focus on, using that very technology to tweet about it. But I think it made a much better world, but it created a lot of new dangers. So maybe you can speak to when you have, when you kind of stand at the end of the road for building a really solid, reliable, universal computer. What are the possibilities that are positive? What are the possibilities that are negative? How can we minimize the chance of the negative? [/INST] Yeah, that's a really good question. So there's so many positive things from drug discovery, from supply chain stress, for basically enabling chemists to basically build more productive in the lab, right? Well, the computer's not going to replace the chemist. There's going to be a Moore's Law of molecules, right? </s><s>[INST] There's going to be so many more molecules we can design, so many more diseases we can cure. So chemists in the lab, as researchers that's better for science. So they can build a bunch of, they can do science at a much more accelerated pace. So it's not just the development of drugs, it's actually like doing [/INST] and the basic understanding and the science of drugs. And the personalization, the cost of drugs right now, we're all living longer, we're all having more and more, we know more about our genomic development, we know about our predetermination, and we might be able to, one dream I've got is like, imagine you can work at your genome assistant, tells you you're gonna get cancer in seven years time, and you have your personal computer that cooks up the right molecule just for you to cure it, right, that's a really positive idea. The other thing is when drugs, so right now I think it is absolutely outrageous that not all of humanity has access to medicine. And I think the computer might be able to change that fundamentally because it will disrupt the way things are manufactured. So let's stop thinking about manufacturing in different factories. Let's say that computers, clinical grade computers or drug grade computers will be in facilities all around the world, and they can make things on demand as a function of the cost, maybe people won't be able to afford the latest and greatest patent, but maybe they'll be able to get the next best thing and will basically democratize, make available drugs to everybody that they need. And there's lots of really interesting things there. So I think that's gonna happen. I think that now let's take the negative. Before we do that, let's imagine what happened to go back to a really tragic accident a few years ago, well, not accident, an act of murder by that pilot on the, I think it was Euro wings or Swiss wings, but what he did is plane took off, he went to his pilot, went to the toilet, he was a co-pilot, he locked the door and then set the autopilot above the Alps, he set the altimeter or the descend height to zero. So the computer just took the plane into the Alps. Now, I mean, that was such a tragedy that obviously the guy was mentally ill, but it wasn't just a tragedy for him, it was for all the people on board. But what if, and I was inspired by this and my thinking, what can I do to anticipate problems like this in the computer? Had the software, and I'm sure Boeing and Airbus will be thinking, oh, maybe I can give the computer a bit more situational awareness. So whenever one tries to drop the height of the plane and it knows it's above the Alps, we'll just say, oh no, computer says no, we're not letting you do that. Of course, he would have been able to find another way, maybe fly it until it runs out of fuel or something, </s><s>[INST] but you know. Keep anticipating all the large number of trajectories that can go negative, all those kinds of running into the Alps and try to at least make it easy [/INST] for the engineers to build systems that are protecting us. Yeah, and let's just think, what in the computer world right now, are CHI-DLs, let's just not think about what I'm doing right now. What I'm doing right now is, it's completely open. Everyone's gonna know CHI-DLs. You can be playing with them, making them easier and easier and easier, but what we're going to start to do, it makes sense to encrypt the CHI-DLs in such a way, let's say you work for a pharmaceutical company, and you have a license to make given molecule. Well, you get issued with a license by the FDA or your local authority and they say, yeah, you're licensed to do it, here, it is, it's encrypted and the Kydial gets run. So you have a license for that instance of use. Easy to do. Computer science has already solved the problem. So the fact that we all trust online banking, right now, then we can secure it. I'm 100% sure we can secure the computer. And because of the way we have a many... It's like the same mapping problem that you... To actually reverse engineer a Kydial will be as hard as reverse engineering the encryption key brute force. It will be cheaper to just actually buy the regulated medicine. And actually people aren't going to want to then make their own fake pharmaceuticals because it'll be so cheap to do it. We'll drop the cost of access to drugs. Now what will happen? Recreational drugs. People will start saying, well, I want access to recreational drugs. Well, it's going to be up to... It's going to accelerate that social discussion that's happening in the US and Canada and the UK everywhere, right? Because cost goes down, access goes up, access. Given cannabis THC to some people who've got epilepsy isn't literally, forgive the term, a no-brainer because these poor people go from seizures every day to maybe seizures just once </s>

<s>[INST] It feels wrong to say, but I was kind of avoided saying it, but if I'm being honest with my feelings, this is a safe space for feeling is I I think it was the greatest mixed martial arts fight I've ever seen and I don't think I'm being biased. I was honestly thinking like, am I being biased? I honestly, I don't think so. I think that was the greatest fighter. Like, if you want to rank fights I've ever seen, I think to me that was the greatest fighter I've ever seen. [/INST] It certainly was one of the greatest displays of like, just dogged effort from an underdog who was out experienced and and probably outsized, but I mean, like you just, Kyle is one of those kids, you're never going to tell him he's out of a fight, he has something you can't teach. And I've seen tons of people with more physical attributes and they're just mental midgets and they got a million dollar body and a 50 cent heart and, and, and Kyle is not that, and you can't teach it no matter what you do, but that was, I would say like my career in combat sports, which spends, you know, if you want to go all the way back to like wrestling, Like that was one of probably the greatest experiences I've I've been a part of it's a bittersweet sport. She's a fickle mistress. </s><s>[INST] Yeah, I mean, the tragic aspect of that is um like, I guess kind of lost, right? So like if you look at the record and all the kind of things, perhaps uh like you look at the career maybe like as a financial, from a financial perspective that perhaps is not uh the greatest thing, forecast career or that or in the in the history of the UFC, perhaps it's not it's not um you know, like maybe many people didn't even watch that fight, but it was a special moment that stands in the history. There's not many of these in uh history of fighting. So, but at [/INST] the end of the day, when you look at someone's career in the UFC, like mhm, financially, there's a, you know, a handful of people that make real money. Everybody else makes nothing. There's a handful of people that make real money. So did that loss cost him in the in the near term. Sure. But when you look back on your life, you're not going to look back on that loss as something that derailed my life financially. And I never recovered from it. That's not gonna happen. Like, the sad thing is unless you were a champion and, you know, most, most people are gonna be forgotten right after they're gone, most people will be forgotten and if you're not forgotten, certainly your accolades are gonna be misrepresented. Either they're going to be inflated or diminished one way or the other. So looking back on it, it's just so hard to, to quantify that. But it's an experience and like when you're in that moment and you're one of the people like intimately involved in it, the value of that experience supersedes any financial gain. </s><s>[INST] Where would you put a bib in the discussion of the greatest of all time. So he, recently we work together, we watched the fight um, of him and um, Justin G and, and could be retired. Would you put him up there as some of the, as one of the greatest or did he never truly find his foil that like the great warrior that challenged him and maybe you want to, do you think he's fully retired now? [/INST] To answer the question about being fully retired? I don't have any idea. I can't for a second pretend to? Um, oh, I think that I understand the way that people from that part of the world think and respect their family and things like that to an american who says, oh, I promised my mom I wouldn't do it. I mean, I promised my mom I wouldn't do a lot of things. I went right out the back door and do them. But I think that that means something different to people in different parts of the world. So I have no idea what kind of weight that carries. So I can't answer that. I can say a lot of times when people think about great fighters, they think about the aspects that make up Emma, like they think of mm A as a pie and there's all these different pieces that make up for make up the pie. And how good is this piece? And how good is this piece? And how good is this piece? When the fact of the matter is you only need one really, really, really good piece and the other pieces are complementary pieces to get you to where you're the strongest. And if you want to tell me that cabin is not the greatest m M A fighter because he doesn't have really slick striking. You can make that argument, but what I can tell you is cupid has good enough striking to get him to his grappling where he is clearly the best guy at 155 they've ever seen. So, does that make him the greatest fighter at that division or not? To your point about the foil. They wanted Connor to be his foil and he just manhandled them. I mean, they wanted that to happen. Did not happen. So </s>

<s>[INST] That's one of the cool things about Ai and robotics or intellectual pursuits or scientific pursuits as you can spend until you're 80 doing it. So I'm in the early days of that one of the reasons I came to texas, one of the reasons I didn't want to pursue an academic career at M. I. T. Is I want to build a company and so that I'm in the early days of that ai company and so it's an open world to see if I'm actually going to be good at it. But the thing that's there that I've been cognizant of my whole life is that I have a passion for it. Something within me draws me to that thing and you have to listen to that to that voice. So with chess your fucked unless you like early on are really training really hard. I think life is more forgiving, you can be, you can be world class at a thing after after making a lot of mistakes and after spending the first few decades of your life doing something completely different in chess, it's like um it's like an olympic sport, like there's no perfection is a requirement as a necessity. What do you think is that pursuit for you? Like why did you decide to stream? What what drew you? [/INST] I like these questions now. We're really getting deep. </s><s>[INST] Yeah, this is like a therapy session. I mean what isn't it terrifying to be in front of a camera? [/INST] Well it's terrifying to be in front of five cameras, corrections </s><s>[INST] six, [/INST] six more </s><s>[INST] terrifying for me to try to remember if I actually turned them all, like I mentioned to you off mike, I'm still suffering from a bit of PTSD after uh screwing up a recording of magnus. I uh had to console me because that was the thing is I felt okay, you wanna you wanna build robots if you can't get a camera to even run correctly, how are you going to do anything else in life? [/INST] Like it </s><s>[INST] was, it was spiraling hard and I was just laying there and just just feeling sorry for myself but I think that feeling by the way and a small tangent is really useful. Like I feel like a lot of growing happens when you feel shitty, as long as long as you can get out of it, Like don't let it spiral like indefinitely. But just feeling really, really shitty about everything in my life. Like I was having an existential crisis like how I how will I be able to do anything at all like this? You're a giant failure, all those kinds of negative voices. Um but I think I made some good decisions in the, in the week after that you [/INST] think you couldn't have made those decisions if you were less hard on yourself? </s><s>[INST] Me personally? No, I'm too lazy. Okay, [/INST] so you really need to be angry at yourself enough to go and do </s><s>[INST] it. It's not even angry, it's just upset of being self critical. [/INST] Like also </s><s>[INST] for me personally, because I don't have Proclivities for depression, I have a lot more room to feel extremely shitty about myself. So if you, if you're somebody that can get stuck in that place like clinically depressed, you have to be really, really careful. You have to notice the triggers, you don't want to get into that place. But for me, just looking empirically feeling shitty has always been productive, like it makes me long term happier. Ultimately it makes me more grateful to be alive. It helps me grow all those kinds of things. So I kind of uh embrace it. Otherwise I feel like I will never do anything, I have to feel shitty. But that's not a thing. I prescribe to others. There's a, there's a famous professor at M. I. T. His name is Marvin Minsky and when he was giving advice about like the students, he said the secret to my success was that I always hated everything I did in the in the past. So always sort of being self critical about everything you've accomplished. Never really take a moment of gratitude. And I think for a lot of people that hear that that's not good, you should be, you should like take a pause and be grateful, but it really works for him. So you know, it's a choice you have to make, [/INST] it reminds me of a quote, be happy, but never satisfied where you can have a positive spin and still want to improve yourself. </s><s>[INST] But yeah, like uh when did you decide to take a step in the spotlight? That terrifying spotlight of the internet? [/INST] It was actually my senior year of college and I was really busy with work and school and chess was kind of like this lost love and the interesting thing is that the longer I don't play chess, the more I kind of miss playing it casually and enjoy it more because then I start looking at it with fresh eyes, but I didn't have time to play tournaments. So I started streaming online because it was more social than just playing strangers on the internet without knowing anything about who they are? And I started slowly growing a community and got in touch with chess dot com pretty quickly too. So then it was this hobby that I would do once a week every thursday at eight p.m. And it was one of the things that brought me a lot of joy and actually I speaking of depression did struggle for with you know, at least 10 years of my life and it was one of those things where chess and streaming was such a distraction and it brought me such great joy that I just kept doing it because I really really liked it. Um And then I was working on something that didn't pan out and decided to go and take a risk and just stream full time, which you know, seemed a little bit weird at the moment, but </s>

<s>[INST] so machine learning, we gotta come up with a better term than programming to appoint our [/INST] But yeah, it's a programming language. </s><s>[INST] Yeah, I wonder if it could be formalized a little bit better, which we feel like we're in the early days what that actually entails [/INST] data driven programming. </s><s>[INST] Data driven programming. Yeah, but so fundamentally different as a paradigm than the others, like it almost requires a different skill set, but you think it's Yeah and pytorch versus tensorflow. Pytorch winds, [/INST] it's the fourth paradigm, It's the fourth paradigm that I've kind of seen there's like this, you know, imperative functional hardware, I don't know a better word for it. And then ml </s><s>[INST] do you have advice? Yeah, people that want to uh you know, get into program. You want to learn programming? You have a video? Uh what is programming? New blessings, exclamation point. And I think the top comment is like warning, this is not for Nubes. Uh do you have a new, like T. L. D. W for that video? But also uh a new but friendly advice on how to get into programming? We're [/INST] never going to learn programming by watching a video called learn programming. The only way to learn programming, I think the only one is the only way everyone I've ever met who can program well, learned it all in the same way they had something they wanted to do and then they tried to do it and then they were like, oh well, okay, this is kind of, you know, be nice if the computer could kind of do this. And then, you know, that's how you learn, you just keep pushing on a project. Um so the only advice I have for learning programming is go program. </s><s>[INST] Somebody wrote to me a question like, we don't really, they're looking to learn about recurring neural networks and saying like, my company is thinking of doing using recurrent neural networks for time series data, but we don't really have an idea of where to use it yet. We just want to like, do you have any advice on how to learn about these are these kind of general machine learning questions and I think the answer is like actually have a problem that you're trying to solve. And and just [/INST] I see that stuff on my growing people talk like that, they're like I heard machine learning is important, Could you help us integrate machine learning with macaroni and cheese production? You just don't even, you can't help these people who lets you run anything, who lets that kind of person run anything. </s><s>[INST] I think we're all uh we're all beginners at some point. So it's [/INST] not like they're a beginner, it's it's like my problem is not that they don't know about machine learning. My problem is that they think that machine learning has something to say about macaroni and cheese production. Like I heard about this new technology, how can I use it for Why? Like I don't know what it is, but how can I use it for why? </s><s>[INST] That's true. You have to build up an intuition of how, because you might be able to figure out a way, but like the prerequisite is you should have a macaroni and cheese problem to solve first. Exactly. And then to you should have more traditional like in the learning process involved more traditionally applicable problems in the space of whatever that is machine learning. And then see if you can be applied to [/INST] at least start with, tell me about a problem? like if you have a problem, you're like, you know, some of my boxes aren't getting enough macaroni in them. Um can we use machine learning to solve this problem? That's much much better than how do I apply machine learning to macaroni and cheese? </s><s>[INST] One big thing, maybe this is me talking to the audience a little bit because I get these days so many messages, a device on how to like learn stuff. Okay, my this this this is not me being mean, I think this is quite a profound actually is you should google [/INST] it. Yeah, </s><s>[INST] like what one of the uh like skills that you should really acquire as an engineer, as a researcher, as a thinker, like 12 complementary skills, like one is with a blank sheet of paper with no internet to think deeply, and then the other is to google the crap out of the questions you have, like that's actually a skill and people often talk about, but like doing research, like pulling at the threads, like looking up different words going into like get hub repositories with two stars and like looking how they did stuff, like looking at the code or going on twitter saying like there's little pockets of brilliant people that are like having discussions, like if you're a neuroscientist going to signal processing community, if you're an ai person going into the of psychology community, like the switch communities that keep searching, searching, searching because it's so much better to invest in like finding somebody else already solved your problem, then there needs to try to solve the problem. And because they've often invested years of their life, like entire communities are probably already out there who have tried to solve your problem. [/INST] I think they're the same thing. I think you go try to solve the problem and then in trying to solve the problem, if you're good at solving problems, you'll stumble upon the person who solved it </s>

<s>[INST] When you say there's a lot of consciousness is is a singular consciousness basically the set of possible experiences. You can have that subjective way as opposed to the the underlying mechanism because you say that, you know having uh extra color receptor uh extra ability to have new experiences. That somehow a different consciousness. Is there a way to see that as all the same consciousness? The subjectivity itself, [/INST] right? Because when we have two of these conscious agents interacting mathematics, they actually satisfy the definition of a conscious agent. So in fact they are a single conscious agent. So so in fact one way to think about what I'm saying that I'm postulating with my colleagues, chetan and chris and others robert partner and so forth. Um There is one big conscious agent in fully complicated But fortunately we can for analytic purposes, break it down all the way to in some sense, the simplest conscious agent which has one conscious experience. one I like this, this one Agent can experience read 35. That's it. That's what, that's what experiences you can get all the way down to that. </s><s>[INST] So you think it's possible the consciousness, whatever that is is uh much more is fundamental or at least much more in the direction of the fundamental than is spacetime as we perceive it. [/INST] That's the proposal. And therefore what I have to do in terms of the hard problem of consciousness is to show how dynamical systems of conscious agents could lead to what we call space and time and neurons and brain activity. In other words, we have to show how you get spacetime and physical objects entirely from a theory of conscious agents outside of space time. With the dynamics outside of space time. So that's that's I can tell you how we plan to do that. But but that's that's the idea. </s><s>[INST] Okay, the magic of it. The chocolate is delicious. So, so there's a mathematical kind of thing that we could say here, how it can emerge within the system of networks of conscious agents. But um is there going to be at the end of the proof why chocolate so delicious or or no, I guess I'm going to ask different kinds of dumb questions to try to sneak [/INST] up. Oh well that's the right question. And when I say that I took conscious experiences as fundamental what that means is in the current version of my theory, I'm not explaining conscious experiences where they came from. That's the miracle. That's one of the miracles. So I have two miracles in my theory, there are conscious experiences like the taste of chocolate and that the there's a probabilistic relationship. When certain conscious experiences occur, others are more likely to occur. Those are the two miracles that </s><s>[INST] possible to get beyond that and somehow start to chip away at the miracle nous of that miracle that chocolate is still delicious. I [/INST] hope so. I've got my hands full with what I'm doing right now. But but the I can just say a top level how we would think about that. That would get at this consciousness without form. This this is gonna be really this is really tough because it's consciousness without form versus the various forms of consciousness takes for the experiences that it has. </s><s>[INST] Right? Right. [/INST] So there's so when I write down a probability space for these conscious experience, I say here's a probability space for the possible conscious experiences, right? It's just like when I write down a probability space for an experiment, like I'm gonna flip a coin twice, right? And I want to look at the probabilities of various outcomes. I have to write down a probability space. There could be heads, heads, heads, tails, tails, heads, tails, tails. So you before as any class in probability, you're told right down your probability space, if you don't write down your probability space, you can't get started. So here's my probability space for consciousness. How do I want to interpret that structure? The structure is just sitting there. There's gonna be a dynamics that happens on it, right. Experiences appear and then disappear just like Heads appears and disappears. So so one way to think about that fundamental probability space is that corresponds to consciousness without any content? The infinite consciousness that transcends any particular content? </s><s>[INST] Well, do you think of that as a mechanism as a thing? Like the rules that govern the dynamics of the thing outside of space time? Isn't that if you think consciousness is fundamental, isn't it? Essentially getting like it is solving the hard problem, which is like from where does this thing pop up, [/INST] which </s><s>[INST] is the mechanism of the thing popping up? Whatever the consciousness is, the different kinds of was so on. That mechanism. And also, the question I want to ask is how how tricky do you think it is [/INST] to </s><s>[INST] solve that problem? You solved a lot of difficult problems throughout the history of humanity, um There's probably more problems to solve left that we've solved by like an infinity. Uh but along the long journey of intelligent species, how when will we solve this consciousness one? Which is one way to measure the difficulty of the [/INST] problem. So I'll give two answers. There's one problem I think we can solve but we haven't solved yet. And that is the reverse of what my colleagues called the hard problem. The problem of how do you start with conscious experiences in the way that I've just described them and the dynamics and build up space and time and brains that I think is a tough technical problem but some principle solvable. So I think we can solve that. So we would solve the hard problem not by showing how brains create consciousness, but how networks of conscious agents create what we call the symbols that we call brains. So that that I think but </s>

<s>[INST] day, someone alive today, but somebody you haven't [/INST] met yet, Who </s><s>[INST] would you be? Would [/INST] I be modeling their brain state or would I just be in their </s><s>[INST] body? [/INST] You can choose </s><s>[INST] the degree to which you are [/INST] modeling their </s><s>[INST] brain state because so you can [/INST] Still take a 3rd person </s><s>[INST] perspective and [/INST] realize you have to realize that </s><s>[INST] your, can they be alive [/INST] or can it be dead? </s><s>[INST] No. [/INST] Oh you they will be brought back to life, </s><s>[INST] right? If they're [/INST] dead? Yeah, you can bring people back definitely Hitler </s><s>[INST] Stalin. [/INST] I want to understand evil. You need to go to </s><s>[INST] experience what it [/INST] feels like. I want to be in their brain feeling what they feel that might change you forever returning from that. Yes, but I think it would also help me understand how to prevent it and fix it. That might be one of </s><s>[INST] those things once you experience [/INST] it will be a burden, </s><s>[INST] you know it [/INST] because you won't be. But </s><s>[INST] a lot a lot of things are our burdens like it's useful but it's a useful burden [/INST] that for </s><s>[INST] sure I want to [/INST] understand evil and like psychopathy and and and that I have all these fake twitter </s><s>[INST] accounts where I like go [/INST] into different algorithmic bubbles to try to like understand. </s><s>[INST] I'll keep getting [/INST] in fights with people and realize we're </s><s>[INST] not actually fighting. [/INST] I think we're we </s><s>[INST] used to exist in a monoculture [/INST] like before social media and stuff </s><s>[INST] like we kind of [/INST] got fed the same thing, so we were all speaking the same cultural language. But I think recently one of the </s><s>[INST] things that like we aren't diagnosing [/INST] properly enough with social media is that um there's different dialects, there's so many different dialects of chinese, there are now becoming different dialects of english, like I am realizing </s><s>[INST] like [/INST] there are people who are saying the exact same things but they're using completely different </s><s>[INST] Burbage and [/INST] we're like punishing each other for not using the correct </s><s>[INST] verbiage and we're completely [/INST] misunderstanding like people are </s><s>[INST] just like misunderstanding [/INST] what the other people are saying and like like I just got in a fight with a friend um about like anarchism and and communism and shipped for like two </s><s>[INST] hours and then [/INST] by the end of the conversation </s><s>[INST] like and then she'd say something [/INST] and I'm like but that's </s><s>[INST] literally what I'm saying and [/INST] she was like what, what? </s><s>[INST] And then and then I was like funk, we have different, [/INST] I'm like we're </s><s>[INST] our english like [/INST] the way we are understanding terminology is like </s><s>[INST] drastically [/INST] like </s><s>[INST] our algorithm bubbles [/INST] are are creating many dialects </s><s>[INST] and and how language [/INST] is interpreted languages </s><s>[INST] used. That's so fascinating. [/INST] And so we're like </s><s>[INST] having these [/INST] arguments that we do </s><s>[INST] not need to be having and [/INST] there's polarization that's happening that doesn't need to be </s><s>[INST] happening because we've got these [/INST] like algorithmic lee </s><s>[INST] created um [/INST] uh dialects </s><s>[INST] occurring plus on top of that there's [/INST] also different parts of the </s><s>[INST] world that speak different languages. [/INST] So there's </s><s>[INST] literally lost [/INST] in translation kind of </s><s>[INST] communication, I [/INST] happen to know the Russian </s><s>[INST] language and I just know how [/INST] different it is. Um, then </s><s>[INST] the english language [/INST] and I just wonder </s><s>[INST] how much is lost in a [/INST] little bit of man. I actually, </s><s>[INST] I have a question for you. I have a song coming out [/INST] tomorrow with, I speak or a Russian band and I speak, I speak a little bit of Russian and I was looking at the title and the title in English doesn't match the title in Russian. I'm curious about this because look, it says </s><s>[INST] title in [/INST] english is last day and then the title in Russian is my practice pronunciation, Sex, no video </s><s>[INST] in like a New day, [/INST] a New day, </s><s>[INST] a new day. [/INST] But yeah, a new day. Uh, yeah, New Day, New </s><s>[INST] Day, but last day, [/INST] uh, nova jeans, So last day will be passed. Maybe they, maybe the title includes both the Russian and the, and the, and it's for maybe maybe for bilingual, be honest, Norwegian sounds better than uh, </s><s>[INST] just um, [/INST] musically like </s><s>[INST] no Virginia's [/INST] New day, that's the current one </s><s>[INST] and [/INST] is the last day. </s><s>[INST] Uh, I [/INST] think Norwegian, but the meaning is so </s><s>[INST] different. [/INST] That's kind of awesome. Actually, there's, there's an explicit set of contrast </s><s>[INST] like that. [/INST] Um, if everyone on </s><s>[INST] earth disappeared [/INST] and it was just you </s><s>[INST] left. [/INST] Um, </s><s>[INST] what would your day look like? Like, what would you do? [/INST] Everybody's dead? Are </s><s>[INST] there corpses there. [/INST] Okay, seriously, it's a </s><s>[INST] big, [/INST] it's a big difference if there's just like birds singing versus if </s><s>[INST] there's like corpses littering [/INST] the streets. Yeah, there's corpses </s><s>[INST] everywhere. I'm sorry. It's and [/INST] you don't actually know what happened and you don't know why you survived </s><s>[INST] and you don't even know if there's others [/INST] out there, </s><s>[INST] but it seems [/INST] clear that it's all gone. What would you do? What would I </s><s>[INST] do? Listen? I'm somebody who really [/INST] enjoys the moment </s><s>[INST] enjoys life. I would [/INST] just go on </s><s>[INST] like [/INST] enjoying </s><s>[INST] the inanimate objects. [/INST] I would just </s><s>[INST] uh [/INST] look for food, basic survival. </s><s>[INST] But most of it is just [/INST] listen, when I just, I </s><s>[INST] take walks and [/INST] I look outside </s><s>[INST] and I'm just happy [/INST] that we get to exist on this planet </s><s>[INST] to be able to breathe air. [/INST] It's just all beautiful. It's full of colors, all of this kind of stuff. Just, there's so </s><s>[INST] many things about life, [/INST] your own </s><s>[INST] life, conscious life. That's fucking awesome. So [/INST] I would just enjoy that. </s><s>[INST] But also, [/INST] maybe after a few weeks </s><s>[INST] the engineer will start [/INST] coming out, like, want to build some </s><s>[INST] things. [/INST] Maybe there's always hope searching for another human, </s><s>[INST] probably searching for [/INST] another human, probably trying to get to a tv </s><s>[INST] or radio station [/INST] and broadcast something I that's interesting. I didn't think about </s><s>[INST] that. So like [/INST] really maximize your ability to connect with others. Yeah. Like probably try to find another person. Would you be excited to </s>

<s>[INST] the togetherness of what, what, who's God? All those gods. The stories have been told throughout history, you said through the spiritual traditions. What do you think that is, is that us trying to find that common thing at the core? Oh, [/INST] well, in in in many traditions, not all The one I was raised in. So my dad was a protestant minister. We tend to think of God as a being, but I think that that's not right. I think the closest way to think about God is being period, not a being, but being the very ground of being itself is God. I think that's the, the deep and from my point of view, that's the ground of consciousness. So the ground of conscious being is what we might call God. But but the word God has always been, you know, for example, you don't believe the same God is my God. So I'm gonna fight you and we'll have wars over because the being, the specific being that you call God is different from the being that I call God. And so we fight whereas if it's not a being, but just being and you and I share being, then you and I are not separate and there's no reason to fight. We're both part of that one being and loving you is loving myself because we're all part of that one being. The spiritual traditions that point to that I think are pointing in a very interesting direction. And that does seem to match with the mathematics of the conscious agent stuff that I've been working on as well that it really fits with that, although that wasn't my goal is </s><s>[INST] there, you mentioned, you mentioned that the young physicist that you talk to or whose work you follow have quite a lot of fun breaking with the traditions of the past, the assumptions of the past. What advice would you give to young people today in high school and college, not just physicists, but uh in general, how to have a career. They can be proud of how they can have a life. They can be proud of how to make their way in the world, from the lessons from the winds and the losses in your own life. What what, what little insights could you pull out? [/INST] I would say the universe is a lot more interesting than you might expect and you are a lot more special and interesting than you might expect? You might think that you're just a little tiny irrelevant, £100, £200 person in a vast billions of light years across space. And that's not the case. You are in some sense the being that's creating that space all the time. Every time you look so waking up to who you really are outside of space and time as the author of Space and Time is the author of everything that you see, </s><s>[INST] the author of Space and Time. You're [/INST] the author of Space and Time and I'm the author of of Space and Time and Space and Time is just one little data structure. Many other consciousnesses are creating other, other data structure, their authors of various other things. So, so realizing and then realizing that that um I had this feeling that growing up and going read all these texts. Oh man, it's all been done. Yeah. If I'd just been there 50 years ago I could have discovered this stuff, but it's all in the textbooks now. Well, Believe me, the textbooks are going to look silly in 50 years and it's your chance, your chance to write the new textbook. So, so of course study the current textbooks. You have to understand that there's no way to progress until you understand what's been done. But then the only limit is your imagination, frankly, that's the only limit, </s><s>[INST] the greatest books, the greatest textbooks ever written on earth are yet to be written. Uh what do you think is the meaning of this whole thing? What's the meaning of life from your limited interface? Can you, can you, can you figure it all out? Like what, why? So you said the universe is kind of trying to figure itself out through through us, uh why? Why? [/INST] That's the closest I've come. So I'll give you so I, I will say that, I don't know, but but here's my guess </s><s>[INST] it's a good first sentence, that's a good starting point [/INST] and and maybe that's going to be profound. Part of the final answer is to start with the, I don't know, it's quite possible that that that's really important to start with the, I don't know, my guess is that if consciousness is fundamental and a girdle and completing this term holds here and there's an infinite variety of structures for consciousness to some sense, explore um that maybe that's what is about this is something that Annika and I talked about a little bit and she doesn't like this way of talking about it and I'm gonna have to talk with some more about this way of talking. But right now I'll just put it this way and I'll have to talk with her more and see if I can say it more clearly. But the way I'm talking about it now is that there's a sense in which there's being and then there's experiences where forms that come out of being. That's one deep deep mystery and the question that you asked, what is it all about? Somehow it's related to that? Why does being why don't you just stay without any forms? Why does it why do we have experiences? Why should why why not just have when you close your eyes and pay attention to what's behind you? There's nothing but there's being why why don't we just stop there? Why didn't we just stop there? Why did we create all tables and chairs in the sun and moon and and people that all this really complicated stuff why and all I can guess right now and I'll probably kick myself in a couple of years and say that was dumb. But but all I can guess right now is that somehow consciousness wakes up to itself by knowing what it's not. So here I am, I'm not this body. I sort of saw that it was sort of in my face when I sent a text goodbye. But then as soon as I'm better sort of like okay, I sort of don't want to go there. Right I okay, so I just so I am my body, my go back to the standard and I am my body and you know, and then I want to get that car and even though I was just about to die a year ago, so that comes rushing back. So so consciousness immerses itself fully into a particular headset gets lost in it and then slowly wakes up, </s>

<s>[INST] DNA data and fitness tracker data to provide you with a clear picture what's going on inside you and to offer you science backed recommendations for positive diet and lifestyle changes. Andrew Huberman loves it and uses it and talks a lot about it. David Sinclair as well in my conversation and in other places talks a lot about it. I feel like too much of health care is based on very limited amount of information that the doctor has to uh make decisions based on very quickly. I think tracking biological data over a prolonged period of time and having that data and making decisions based on that data is just such a more powerful way to make good decisions that are personalized to you For a limited time, you can get 25% off the entire inside track or store if you go to inside tracker.com slash flex. That's inside Tracker dot com slash legs. The show is also brought to you by net suite. That's what allows you to manage financials, HR inventory, e commerce and many more business related details all in one place, running a company is really difficult. But ultimately for most people, I think work is a source of meaning a source of happiness. So I see running a company as not just a place to sort of achieve productivity but also a [/INST] place to achieve </s><s>[INST] happiness for the people involved and that means using the best tools for the job in managing all aspects of that company. That's where all the sort of complex things like HR and inventory tracking all come in and that's where nets, which is a great tool to help you out on that front. Anyway right now, special financing is back at the net suite dot com slash let's get their one of a kind financing program that's not sweet dot com slash lex if you're running a business. I wish you luck my friends. It's a heck of a journey. It's a tough journey. But I think it's a beautiful one. This show is also supported by ladder. I often meditate on my mortality. I think this is both honest and clarifying if you have others in your life you love and you do the same, you should consider getting life insurance ladder can help you with that pay a little every month in case something happens to you and the ones you love will be protected. Like I said, one of the most powerful tools from the book of the stoics is to meditate on your mortality daily to remind yourself that this ride is over. Soon to uh come face to face with that reality that really makes the moment, the minute the hour, the day taste sweeter. Anyway, go to ladder life dot com slash lex to see if you're instantly approved. That's ladder spelled L A D D er life dot com slash lex And may I humbly suggest that at this moment you take one breath in, one breath out and remember that life is finite and you may even die today. So make the most of it. This show is also brought to you by sun basket for some basket delivers fresh, healthy delicious meals straight to your door, starting at only 8 99 per meal. They have prepared meals, meal kits, raw ingredients for me eating is a source of joy. The social aspect, but also just the culinary like I said, uh sophistication, the opportunity to explore different taste. That said, I'm pretty minimalistic. I'm pretty sort of have a driven in the meals I eat, I eat basically the same thing every day. So sun basket is a great thing for me to keep things healthy to keep things efficient, but also to be able to eat all kinds of variety of foods. Anyway, right now, son basket is offering $90 off for your first four deliveries, including free shipping on the first box. When you go to some basket dot com slash lex and to code lex, that's sun basket dot com slash lex and to code lex everything they have is quite honestly healthy and delicious. This is the lex Fridman podcast and here is my conversation with sean kelly. Your interests are in post content, european philosophy, especially phenomenology and existentialism. So let me ask what to you is existentialism. [/INST] So it's a hard question I'm teaching a course on existentialism right now. You are, I am, yeah, existentialism in literature and film, which is fun. I mean, the traditional thing to say about What existentialism is is that it's a movement in mid-20th century, mostly French some German philosophy and some of the major figures associated with it are people like jean paul, sartre and camus um surrender Beauvoir, maybe martin Heidegger, but that's a weird thing to say about it because most of those people denied that they were existentialists. And um and in fact, I think of it as a movement that has a much longer history. So when I try to describe what the core idea of existentialism is, it's an idea that you find expressed in different ways in a bunch of these people. One of the ways that it's expressed is that Start will say that existentialism is the view that there is no god and at least his form of existentialism, he calls it atheistic existentialism, there is no god. And since there is no God, there must be some other being around who does something like what God does otherwise, there wouldn't be any possibility for significance in life and that being is us. And the feature of us according to start and the other existentialists, that puts us in the position to be able to play that role is that we are the beings for whom as starts as it existence precedes essence. That's that's the catch phrase for existentialism. And then you have to try to figure out what it means </s>

<s>[INST] Yeah. So okay, so the brain is a meta learning system with a bunch of, with reinforcement learning agents and what I think you said. But just to clarify, where do the, there's no centralized uh government that tells you Here's a loss function. Here's a loss function. Here's a loss function like what who is who says what's also [/INST] governments which impose lost functions on different parts of the brain. So we have differential attention. Some areas in your brain gets especially rewarded when you look at faces. If you don't have that you will get postive agnosia which basically the inability to tell people apart by their faces. So </s><s>[INST] and the reason that happens is because it was had an evolutionary advantage like evolution comes into play here about [/INST] it's basically an extraordinary attention that we have four faces. I don't think that people were supposed to have a perceived effective brain. The brain just has an average attention for faces. So people were supposed to have nausea, don't look at faces more than they look at cups. So the level at which they resolve the geometry of faces is not higher than the one that than four cups. And people that don't have prostate agnosia look uh objectively it faces right. For you and me. It's impossible to move through a crowd without scanning the faces. And as a result, we make insanely detailed models of faces that allow us to discern mental states of people. </s><s>[INST] So obviously we don't know 99% of the details of this meta learning system. That's our mind. Okay. But still we took a leap from something much dumber to that from through the evolutionary process. Can you first of all maybe say how hard do you? How big of a leap is that from our brain? From our ape ancestors to multi cell organisms? And is there something we can think about about? As we start to think about how to engineer intelligence? Is there something we can learn from evolution [/INST] in some sense, life exists because of the market opportunity of controlled chemical reactions? We compete with dumped chemical reactions and we win in some areas against this damn combustion because we can harness those entropic gradients where you need to add a little bit of energy in a specific way to harvest more energy. So we </s><s>[INST] are competed [/INST] combustion in many regions we do and we try very hard because when we are in direct competition we lose, right? So because the combustion is going to close the entropy gradients much faster than we can run </s><s>[INST] it. [/INST] So yeah. So basically we do this because every cell has a touring machine built into it. It's like literally read write head on the tape. And so everything that's more complicated than a molecule that just is a vortex around the tractors, uh that needs a turing machine it for its regulation. And then you find cells together and you get next level organization organism where the cells together implement some kind of software. And uh for me, very interesting discovery in the last year was the word spirit. Because I realized that what spirit actually means. It's an operating system for an autonomous robot. And when the word was invented, people needed this word, but they didn't have robots that they built themselves. Yet. The only autonomous robots that were known where people, animals, plants, ecosystems, cities and so on. And they all had spirits. And it makes sense to say that the plant is an operating system, right? If you pinch the plant in one area, then is going to have repercussions throughout the plant. Everything in the plant is some sense connected into some global aesthetics like and other organisms. An organism is not a collection of cells is a function that tells cells how to behave. And this function is not implemented as some kind of supernatural thing, like some more for genetic field, it is an emergent result of the interactions of each service each other cell. </s><s>[INST] Right. So [/INST] you're, what you're saying is </s><s>[INST] the organism is a function that tells what's what, what [/INST] tell us what to do. </s><s>[INST] And the function is an image emerges from the interaction of the cells. [/INST] So it's basically a description of what the plant is doing in terms of macro states and the micro states, the physical implementation are too many of them to describe them. So that uh software that we use to describe what the plant is doing. The spirit of the plant is the software the operating system of the plant. Right. This is a way in which we, the observers make sense of the plant and the same is true for people. So people have spirits which is their operating system in a way rightness aspects of that operating system that relate to how your body functions and others how you socially interact, how you interact with yourself and so on. And we make models of that spirit and we think it's a loaded term because it's from a pre scientific age. But we it took the scientific age a long time to rediscover a term that is pretty much the same thing. And I suspect that the difference is that we still see between the old word in the new world translation arrows that over the centuries, </s><s>[INST] can you actually look around that? Like, well why do you say the spirit just to clarify? Because I'm a little bit confused. So the word spirit is a powerful thing. But why did you say in the last year? So that you discovered this, uh do you mean the same old traditional idea of a spirit? Or do you [/INST] try to find out what people mean by spirit? When people say spirituality in the US, it usually refers to the phantom limb that they developed in the absence of culture. And a culture is, in some sense, you could say the spirit of a society that is a long game. The thing that is, becomes self aware at a level above the individuals where you say if you don't do the following things, then the grand grand grand grand Children of our Children will not have nothing to eat. So if you take this long scope where you try to maximize the length of the game that you're playing as a species, you realize that you're part of the largest thing that you cannot fully control, You probably need to submit to the equal sphere instead of trying to completely control it, right? There needs to be a certain level at which we can exist as a species if you want to endure and our culture is not sustaining this anymore. We basically made this bet with the industrial revolution that we can control everything. And the modernist societies was basically unfettered growth led to a situation in which we depend on the ability to control the entire planet. And since we're not able to do that as it seems, this cultural will die and we realized that it doesn't have a future would be called our Children generations. That it's sort of very optimistic thing. </s>

<s>[INST] So that's really interesting. So, you've, you've talked about agency, I think maybe calling it a convenient fiction at all these different levels. So even at the human individual level, it's kind of a fiction. We all believe because we are like you said, made of cells and cells are made of atoms. So that's a useful fiction. And then there's nations that seems to be a useful fiction, but it seems like some fictions are better than others. You know, there's a lot of people that argue the fiction of of nation is a bad idea. One of them lives two doors down from me michael malice. He's an anarchist. You know, I'm sure there's a lot of people who are into meditation that believe the idea this useful fiction of agency of an individual is troublesome as well. We need to let go of that in order to truly like to transcend. I don't know, I don't know what, what words you want to use, but suffering or to, uh, to elevate the experience of life. So you're kind of arguing that, okay, so we have some of these useful fictions of agency. We should add a stronger fiction that we tell ourselves about. The agency of groups in the hundreds of the half of Dunbar's number five X. Dunbar's number [/INST] Yes. Something on that order. And we call them fictions. But really the rules of the game, right? Rules that we we, we feel are fair, our rules that we consent to. </s><s>[INST] I always question the rules when I lose like a monopoly. That's when I usually question when I'm winning. I don't question the rules. [/INST] We should play a game monopoly someday. There's a trippy version of it that we could do </s><s>[INST] what, what kind of [/INST] contract monopoly is introduced by a friend of mine. To me where you can write contracts on future earnings or landing on various things and you can hand out like, you know, you can land the first three times you land on park, place is free or whatever just, and then you can start trading those contracts for money </s><s>[INST] and then you create human civilization. Uh, and somehow, Bitcoin comes into it. Okay. Uh, but some of these [/INST] actually, I bet if me and you, and ERIC sat down to play a game of monopoly and we were to make N. F. T. S Out of the contracts we wrote we could make a lot of money. Now it's a terrible idea. I would never do it. But I bet we could actually sell the N. F. T. S around. </s><s>[INST] I have other ideas to make money that I could tell you. And they're all terrible ideas um including cat videos on the internet. Okay. But some of these rules of the game, some of these fictions are it seems like there better than others. [/INST] They have worked this far to cohere human to organize human collective action. </s><s>[INST] But you're saying something about especially this technological age requires modified fictions. Stories of agency. Why the Dunbar number. And also, you know, how do you select the group of people? You know, Dunbar numbers. I think I have the sense that it's overused as a kind of law that somehow we can have deep human connection at this scale. Like some of it it feels like an interface problem too. It feels like if I have the right tools, I can deeply connect with a large number of larger number of people. It just feels like there's a huge value to interacting just in person getting to share traumatic experiences together. Beautiful experiences together. But there's other experiences like um that in the digital space that you can share. It just feels like Dunbar's number can be expanded significantly. Perhaps not to to the level of millions and billions but it feels like it could be expended. So how do we find the right interface you think? Um for having a little bit of a collective here that has agency, [/INST] you're right that there's many different ways that we can build trust with each other. Um my friend joe Edelman talks about a few different ways that um you know, mutual appreciation, trustful conflict, um just experiencing something like, you know, there's a variety of different things that we can do, but all those things take time and you have to be present, the less present you are. I mean there's just again a no free lunch principle here, the less present you are, the more than you can do, but then the less the last connection you built. So I think there is sort of a human capacity issue around some of these things now that being said yes, we can use certain technologies. So for instance if I write a little monograph on my view of the world, you read it a synchronously at some point and you're like wow, peter, this is great. Here's mine. I read it, I'm like wow lex, this is awesome. We can be friends without having to spend 10 years, you know figuring this stuff out together. We just read each other's thing and be like, oh yeah, this guy is exactly in my wheelhouse and vice versa and we can then um you know, connect just a few times a year and maintain a high trust relationship. It can expand a little bit but it also requires these things are not all technological nature requires the individual themselves to have a certain level of capacity To have a certain lack of neuroticism, right? If you want to use like the the ocean big five sort of model people have been pretty centered, the less centered you are the fewer authentic connections you can really build for a particular unit of time. It just takes more time. Other people have to put up with your crap. Like there's a lot of the stuff that you have to deal with if you are not so well balanced. Right? So yes we can help people get better to where they can develop more relationships faster and then you can maybe expand Dunbar number by quite a bit but you're not going to do it. I think it's really hard to get it beyond 10 X. Kind of the rough swag of what it is, you know. </s>

<s>[INST] no, I I mean part of it was a joke, but as I was saying it [/INST] that that </s><s>[INST] also comes out in the in the book because they're both fucked. They're both they're both um have no control of their fate. Um That's the same was true in um the camp guards in Nazi Germany and and the people in the camps, the worst was brought out in the guards. But there are in it together in some dark way. They're both fucked by a very powerful system that put them in that place. And both of us could be either player in that system, which is the dark reality that Solzhenitsyn also reveals that the line between good and evil runs through the heart. Every man as he wrote in Gulag archipelago. But it is that amidst all of that there's a, I don't know, the good vibes, the positivity comes out from the both of you and that's beautiful to see that is I suppose friendship? What do you think makes a good friend? [/INST] Oh God. I mean it's Bill, you know, it's a billion, it's a billion things that make a good friend. But I think you could break it down to some Rgb. I think you can go Rgb with like a good friendship. Yeah. Yeah, I think you could break up with some like fundamental qualities of friendship and I'd say uh number one, it's love, like it's friendship is love, it's it's a form of love. Uh So obviously without that, I don't know how you I mean, I'm not saying, I think if your true friends you love each other so you need that but love obviously it's not that's not that's not enough. It's like with uh true friends have to be like incredibly honest with each other and not like you know what I mean? But not like uh I don't like I think there's a kind of like, I don't know if you've ever noticed like some people who say, you know, I just tell it like it is but the thing that yeah, why is it that you're telling like it is is always negative. Why is it? It's always cynical or shitty or you're like nagging somebody or me? How come you're not telling it? Like it is when it's good to see, you know what I mean? So sort of like trust, but but a pro evolutionary kind of trust, you know what I mean? Like, you know that your friend loves you and wants you to be yourself because if you weren't yourself, then you wouldn't be their friend, you'd be some other thing. But also they might be seeing your blind spots that other people in your life, your family, your wife, whoever might not be seeing. So that's a good friend is someone who like loves you enough to when it matters be like, hey, are you all right? And then help you see something you might not be seeing, but hopefully they only do that once or twice a year. Yeah. There's </s><s>[INST] something, I mean, it just would have this world, especially in if you're a public figure that this, this world has its, um, has plenty of critics and it feels like a friend, the criticism part, it's already done for you. I think a good friend is just there to support to actually notice the good [/INST] stuff. But you in comedy we need like that. What what like, it's really good in comedy to have somebody you can like be like, what do you think of that and know that they're not going to be like that. </s><s>[INST] But that's for the craft, the craft itself. Like the work you do not the uh, yeah, interesting. But that's so tough. [/INST] Yeah. Whatever your particular art form or whatever you are doing. I mean you don't always be leaning on your friends opinions for like your own innovation, but it's nice to know that you have someone who, not just with jokes, but with anything, If you go to them and run something by them, they're gonna like, they're gonna be honest with you about, like, their real feelings regarding that thing because it helps you grow as a person. We need that and it hurts sometimes and we don't want to hurt our friends. One of the more satanic, like impulses when you're with somebody is, is not wanting to honestly answer whatever they're asking in that regard or wanting to like, put their temporary feelings over something that you've recognized is maybe not great. I'm not, I'm not saying a friendship is something where you're always critiquing or evolving each other. It's not your therapist or whatever. But it's nice when it's there. You know, I think that's another aspect of friendship. </s><s>[INST] Yeah. But yeah, love is at the core of that. You noticed I've met people in my life were almost immediately. Sometimes it takes time where you notice like there's a magic between the two of you. Like, oh, sh it you you seem to be made from the same cloth. [/INST] Yeah, </s><s>[INST] whatever that is, Well, [/INST] you know, we have a name for that and the spiritual community. It's called Sad song and it's, I love the idea. It's uh, basically like if Nietzsche's idea of infinite recurrence is true, then your thought song would be the people you've been infinitely recurring with and those are the people where you run into them and you've never met them, but it's like you're picking up a conversation that </s><s>[INST] I never had [/INST] that and that that, you know, that is based on an idea of like, this isn't the only life is we're always hanging out together, we always show up together, </s>

<s>[INST] So can we, can we try to reverse engineer that math, how that maps to the actual hours. So there's the hours connected to driving are so widely dispersed. As you said, some of them don't count as actual work. Some of it does. That's a very interesting discussion that will then continue when we start talking about autonomous trucking. But uh, you know, you're saying all these cents per mile kind of thing, What uh, how does that map to like our average hourly wage? [/INST] Yeah, so I mean, and this is kind of the, this is also an interesting technology story in the end and it's the technology story that didn't happen. Um, so pay per mile was, you know, invented by companies when you couldn't surveil drivers, you know what they were doing, right? And you wanted them to have some skin in the game. And so you'd say, you know, here's the load it's going from, you know, for me, I might start in, you know, the northeast maybe in upstate new york with a load of beer and say here's this load of beer, bring it to this address in michigan. We're gonna pay you by the mile, right? If I was being paid by the hour, I might just pull over at the diner and and have breakfast. So you're paid by the mile, but increasingly over time, the, the typical driver is spending more and more time doing non driving tasks, there's lots of reasons for that. One of which is railroads captured a lot of freight that goes long distances Now, another one is traffic congestion. Um, and the other one is that drivers are pretty cheap and they're, they're almost always the low people on the totem pole in some segments. And so their time is used really inefficiently. Um, so I might go to that brewery to pick up that load of, of bud light and you know, there, doc staff may be busy loading up five other trucks and they'll say, you know, go over there and sit and wait. We'll call you on the CB when the doctor's ready. So you wait there a couple hours They bring in, you know, you never know what's happening in the truck. Sometimes they're loading it with a fork lift, maybe they're throwing 14 pallets on their full of kegs, but sometimes it'll take them hours, you know, and you're sitting in that truck and you're essentially unpaid, you know, um, then you pull out, you've got control over what you're gonna get paid based on how you drive that load. Um, and then on the other end, you got a similar situation of, of kind of waiting. So, </s><s>[INST] so if that's the way truck drivers are paid and there's a low incentive for the optimization of the supply chain to make them more efficient, right to, uh, to utilize truck labor more efficiently. Absolutely. And so that's the technology problem that uh one of several technology problems that could be addressed. Um I mean what, so what did uh if we just linger on it, what are we talking about in terms of dollars per hour? Is it close to minimum wages? You know, there's something you talk about. There was uh a conception or misconception that truckers get paid a lot for their work. Do they get paid a lot for their work? [/INST] Some do. Uh And I think that's part of the complexity. So you know, what interested me as an ethnographer about this was, you know, I'm interested in the kind of economic conceptions that people have in their heads and and how they lead to certain decisions in labor markets, you know why some people become an entrepreneur and and other people become a wage labor or you know why some people want to be doctors and other people want to be truck drivers. That conception right, is getting shaped in these labor markets is the argument of the book And the fact that drivers can hear or potential drivers can hear about these, you know, workers who make $100,000 plus, which which happens regularly in the trucking industry. There are many truck drivers who make more than $100,000 a year. Um you know, is an attraction, but the industry is highly segmented. Um And and so the entry level segment and and we can probably get into this, but, you know, the industry is dominated by uh, you know, a few dozen really large companies that are self insured and can train new drivers. So if you want those good jobs, you've got to have several years up until recently, the labor markets becoming tighter, but you had to have several years of accident free, you know, perfectly clean record driving to get into them. The other part of the segment, you know, those drivers often don't make minimum wage, but this leads to one of the sort of central issues that has been in the courts and in the legislature, um, in in some states Is, you know, what should truck drivers get paid for? Right? The industry, you know, for the last 30 years or so has said essentially it's the hours that they log for safety reasons for the Department of Transportation, right? Um Now, since the drivers are paid by the mile, They try to minimize those because those hours are limited by the federal government. So the federal government says, you can't drive more than 60 hours in a week as a long haul truck driver. And so you want to drive as many miles as you can in those 60 hours. And so you under report them, right? And so what happens is the companies say, Well, that guy, you know, he only said he logged 45 hours of work that week or 50 hours of work, that's all we have to pay him minimum wage for um when in fact typical truck driver in these and these jobs will work according to most people would sort of define it as like okay I'm at the customer location, I'm waiting to load and doing some paperwork, you know um inspecting the truck, I'm feeling it Um just waiting to you know get put in the dock 80-90 hours would be sort of a typical work week for one of these these drivers um just look at that, they don't make minimum wage often times </s>

<s>[INST] Yeah, but then you operate in the secrecy and then there's there's uh corrupt in douchebag people everywhere. So when they even inside the C. I. A. And criminals inside the CIA, there's criminals and all organizations in all walks of life. Human nature is such that this is always the case. Then it breeds conspiracy theories. Um [/INST] and sometimes those conspiracy theories turn out to be true. But most times they don't that's just part of the risk of being a myth. Can </s><s>[INST] you speak to some of the myths? So, MK ultra, So [/INST] not a myth, </s><s>[INST] not a myth. So this is a fascinating human experimentation program undertaken by the CIA to develop procedures for using drugs like LSD to interrogate people through, let's say psychological manipulation and maybe even torture. The scale of the program is perhaps not known. How do you make sense that this program existed [/INST] again? You've got to look through the lens of time, you've got to look at where we were historically at that time there was the peak of the Cold War. Our enemies were doing the same kind of experimentation. It was essentially of another space race. What if they broke through a new weapon technology faster than we did. What would that mean for the safety and security of the american people. So right decision or one wrong decision. It was guided by an informed by national security priorities. So from this program that was designed to use drugs to drive interrogation and torture people was born something very productive operation Stargate. Which was a chance to use remote viewing and metaphysics to try to collect intelligence. Now, even though in the end the outcome of MK Ultra and the outcome of Stargate were mixed. Nobody really knows if they did or didn't do what they were supposed to do. We still know that to this day. There's still a demand in the U. S. Government and in C. I. A. For people who have sensitivities to ethereal energies </s><s>[INST] by the way, is there any [/INST] proof that that kind </s><s>[INST] of stuff works? Or is it just [/INST] shows it </s><s>[INST] shows that there's interest? It shows that there's openness to consider those kinds of things. But is there any evidence that that kind of stuff works? [/INST] If there's evidence? I haven't seen it. Speaking from a science based point of view, only if energy and matter can always be exchanged, then a person who can understand and and become sensitive to energy is a person who could become sensitive to what does become matter. </s><s>[INST] Yeah. I mean the basics of the physics might be there. But a lot of people probably are skeptical. [/INST] I'm skeptical too. But I'm just like you should be </s><s>[INST] open minded. That's that's that's actually, you know, that's what science is about is remain open minded even for those things that are long shots because those are the things that actually defines scientific revolutions. What about Operation Northwoods? It was a proposed 1962 false flag operation by the D. O. D. And the C. I. A. To be carried out by the C. I. A. To commit acts of terrorism on americans and blame them on CUBA. So JFK the president rejected the proposal. [/INST] What </s><s>[INST] do you make that? This was on the table. Operation Northwoods. So [/INST] it's interesting. Uh, first I'm glad that JFK rejected it. That's uh, that's a good sign. So we have to understand that good ideas are oftentimes born from bad ideas. Uh, I had a really good friend of mine who actually went on to become a pastor and he used to say all the time that he wanted all the bad ideas on the table. Like give me all your bad ideas every time we had any kind of conversation. And I was always one of those people who like isn't a bad idea. Just a waste of time. And he was like, no, because the best ideas oftentimes come from bad ideas. So again, Cuban missile crisis of mass hysteria in the United States about nuclear war from cuba missiles blowing up american cities faster than we could even see them coming. It makes sense to me that a president would go to, especially the part of C. I. A. Which is the special activities division. It makes perfect sense to me that the president would go to a division called special Activities whose job it is to create. You know crazy ideas that have presidential approval but nobody knows they exist. So it makes sense that he would challenge a group like that to come up with any wacky idea. Right? Come up with anything. Just let's start with something because we can't have we can't bring nothing to the table. We have to do something about this Cuban issue. And then that's how an operation like that could reasonably be born. Not because anybody wants to do it but because they were tasked by the president to come up with five ideas and it was one of the ideas </s><s>[INST] that [/INST] still happens to this day. The president will still come in but will basically send out a notice to his covert action arm and he will say I need this and I need it on Wednesday. And people have to come back with options for the thing. He asked for a finding, he will issue a presidential finding and then his covert action arms have to come back and say here's how we would do this and hide the hands of the americans. </s><s>[INST] How gangster was it of JFK to reject it though [/INST] this baller </s><s>[INST] right? That's [/INST] like that is the that is a mic drop right there, nope not doing that, doing that, </s>

<s>[INST] that's almost like creating an internet. I mean that's what google did for web pages that did, uh, you know, page rank, search discovery. You don't, you don't follow anybody on google when you use a search engine, you just discover web pages. And so what Tiktok does is saying, let's start from scratch. Let's like, like start a new internet and have people discover stuff on that new internet within a particular kind of pool of people. [/INST] What's so fascinating about this is like the field of information retrieval, like always talked about as I was studying this stuff, we always use the word query in document. So I was like, why are they saying aquarium documents? Like they're literally imagine like if you just stop thinking about query is like literally a search query and inquiry could be a person. I mean a lot of the way I'm not going to claim to know how instagram or facebook machine learning works today. But you know, if you want to find a match for a query, the query is actually the attributes of the person their age, their gender where they're from maybe some kind of summarization of their interests and that's a query, right? And that matches against documents and by the way, documents don't have to be text. They can be videos there, however long, I don't know what the limit is on Tiktok these days. They keep changing it. My point is just, you've got a query which is someone in search of something that they want to match and you've got the document, it doesn't have to be text, it could be anything. And how do you match make? And that's one of these, like, I mean I've spent a lot of time thinking about this and I don't claim to have mastered it at all, but I think it's so fascinating about where that will go with new social networks. </s><s>[INST] See what I'm also fascinated by is metrics that are different than engagement. So the other thing from an alien perspective, what social networks are doing is they, they in the short term bring out different aspects of each human being. So first let me say that an algorithm or a social network for each individual can bring out the best of that person or the worst of that person or there's a bunch of different parts to us. Parts. We are proud of that, that we are parts we're not so proud of When we look at the big picture of our lives when we look back 30 days from now, am I proud that I said those things are not my problem that I felt those things and my problem that I experienced or read those things or thought about those things just in that kind of self reflective kind of way and so coupled with that I wonder if it's possible to have different metrics that are not just about engagement but are about long term happiness growth of a human being when they look back and say I am a better human being for having spent 100 hours on that app. And that feels like it's actually strongly correlated with engagement in the long term, in the short term. It may not be, but a long term, it's like the same kind of thing where you really fall in love with the product, you fall in love with an iphone, you fall in love with the car. That's what makes you fall in love is like really being proud and just in a self reflective way, understanding that you're better human being for having used the thing and that's like where uh, that's a great relationships for me from it. It's not just like you're hot and we like being together or something like that. It's more like I'm a better human being because I'm with you and that feels like a metric that could be optimized for by the algorithms. Um, but you know, anytime I kind of talked about this with anybody, they seem to say, yeah, okay, that's going to get out competed immediately by the engagement if it's ad driven especially, I I just don't think so. I don't, I mean a lot of it's just implementation [/INST] say a couple of things. One is to pull back the curtain on daily meetings inside of these large social media companies, a lot of what management or at least the people that are tweaking these algorithms spend their time on our trade offs. And there's these things called value functions which are like, okay, we can predict the probability that you'll click on this thing or the probability that you'll share it or the probability that you will leave a comment on it or the probability you'll dwell on it. Individual actions, Right? And you've got this neural network that basically has a bunch of heads at the end and all of them are between zero and one and great. They all have values right? Or they all have probabilities. Um, and then in these meetings, what they will do is say, well, uh, how much do we value a comment versus a click versus a share versus uh, and then maybe even some downstream thing. Right. Um, that has nothing to do with the item there. But like driving follows or something. And what typically happens is they will say, well, what are our goals for this quarter at the company? Oh, we want to drive, sharing up. Okay, well let's turn down these metrics and turn up these metrics and and they blend them right into a single scaler with which they're trying to optimize. That is really hard because invariably you think you're solving for, I don't know something called meaningful interactions, right. This was the big facebook pivot and I don't actually have any internal knowledge, like I wasn't in those meetings, but at least from from what we've seen over the last month or so it seems by actually trying to, trying to optimize for meaningful interactions. It had all these side effects of optimizing for these other things. Um, and I don't claim to fully understand then, but what I will say is that trade offs abound and as much as you'd like to solve for one thing, if you have a network of over a billion people, you're going to have unintended consequences either way and it gets really hard. So what you're describing is effectively a value model that says like can we capture this is the thing that I spent a lot of time thinking about, can you capture utility in a way that like actually measures someone's happiness that isn't just a, what do they call it? A surrogate problem where you say, well I kind of think like the more you use the product, the happier you are, that was always the argument at facebook by the way, it was like, well people use it more. So there must be more happy turns out there like a lot of things you use more that make you less happy in the world not talking about facebook, just you know, let's think about whether it's gambling or whatever like that you can do more of, but doesn't necessarily make you happier. So the idea that time equals happiness, obviously you can't map utility and time together easily, There are a lot of edge cases. So when you look around the world and you say well what are all the ways we can model utility? There is like one of the please if you know someone smart doing this introduced me because I'm fascinated by it and it seems really tough but the idea that reinforcement learning like everyone interesting I know and machine learning, like I was really interested in recommend er systems and supervised learning and and the more I dug into it was like literally everyone smart is working on reinforcement learning, like literally everyone, </s>

<s>[INST] Yeah. Do you, can you think about that for a little bit? Uh, do you miss those experiences? Because it's a fundamentally different sets of experiences that you no longer have? Or do you or is it just a nice thing to have had? You don't see them as that fundamentally different than you visiting a new country and experiencing new environments? [/INST] I guess for me when I had these experiences, they were somewhat marginal. They were like a little bonus kind of experience. I know there are people who have much more serious forms of synesthesia than this for whom it's absolutely central to their lives and people who, when they experience new people, they have colors, Maybe they have tastes and so on. Every time they see writing, it has, uh, it has colors. Some people whenever they hear music, it's gotta, it's got a certain really rich color pattern and for some synesthesia is absolutely essential. And I think if they lost it, they'd be devastated again. For me it was a very, very mild form of synesthesia in it's like, yeah, it's like those interesting experiences. Um you know, you might get under different auto states of consciousness and so on. It's kind of cool, but not necessarily the single most important experiences in your life. </s><s>[INST] Got it. So let's try to go to the very simplest question, the events of any time. But perhaps the simplest things can help us reveal even in time, some some new ideas. What in your view is consciousness? What is quality to what is the hard problem of consciousness? [/INST] Consciousness? I mean, the word is used in many ways, but the kind of consciousness that I'm interested in is basically subjective experience what it feels like from the inside to be a human being or any other conscious being. I mean, there's something it's like to be me right now, I have visual images that I'm experiencing. I'm hearing my voice. I've got maybe some emotional tone. I've got a stream of thoughts running through my head. These are all things that I experienced from the first person point of view of sometimes called this the inner movie and the mind sort of perfect. It's not a perfect metaphor. It's not like a movie in every ways and in every way and it's very rich. But yeah, it's just direct subjective experience. And I call that consciousness or sometimes philosophers use the word quagliana, which you suggested. People tend to use the word qualify for things like the qualities of things like colors, Redness, the experience of redness versus the experience of greenness. The experience of one taste or one smell versus another. The experience of the quality of pain. And a lot of consciousness is the experience of those of those are those qualities </s><s>[INST] of consciousness is bigger. The entirety of any kind of [/INST] consciousness of thinking is not obviously Quaglia, it's not like specific qualities like redness or greenness. But still, I'm thinking about my hometown. I'm thinking about what I'm gonna do later on. Maybe there's still something running through my head which is subjective experience. Maybe it goes beyond those qualities. Or Quagliana philosophers sometimes use the word phenomenal consciousness for consciousness in the sense that many people also talk about access consciousness. Being able to access information in your mind, Reflective consciousness. Being able to think about yourself. It looks like the really mysterious one. The one that really gets people going is phenomenal consciousness. The fact that all of this, the fact that their subjective experience and all this feels like something at all. And then the hard problem is, how is it that why is it that there is phenomenal consciousness at all? And how is it that physical processes in the brain? I could give you subjective experience, it looks like on the face of it. You can have all this big complicated physical system in the brain running without giving subjective experience at all. And yet we do have a subjective experience. So the hard problem has just explained that. </s><s>[INST] Explain how that comes about. We haven't been able to build machines, work a red light goes on, that says it's not conscious. So how does how do we actually create that? Or how do humans do it? And how do we ourselves do [/INST] it? We do every now and then create machines that can do this. We create babies. Yes, that are that are conscious. They've got these brains and brain does produce consciousness. But even even though we can create it, we still don't understand why it happens. Maybe eventually we'll be able to create machines. Which as a matter of fact ai machines, which as a matter of fact, our conscious but that won't necessarily make the hard problem go away any more than it does with babies, because we still one of them. How and why is it that these processes give your consciousness? You </s><s>[INST] know, you just made me realize for for a second, maybe it's a totally dumb realization, but nevertheless, that it's a useful way to think about the creation of consciousness is looking at a baby. So that there's a certain point at which that baby is not conscious sort of uh baby starts from maybe I don't I don't know from a few cells, right? There's a certain point in which it becomes consciousness arrives. It's conscious, of course, we can't know exactly online, but it's a useful idea that we do we do create consciousness again, a really dumb thing for me to say. But not until now that I realized we do engineer consciousness. We we get to watch the process happen. We don't know which point it happens or where it is, but, you know, we do see the birth of consciousness. [/INST] Yeah, I mean, there's a question, of course, is where the babies are conscious when they're born, and it used to be, it seems at least some people thought they weren't, which is why they didn't give anesthetics to newborn babies when they circumcise them. And so now people think, oh, that's incredibly cruel. Of course, of course, babies feel pain. And now the dominant view is that the babies can feel pain. Actually, my partner of claudia, works on this whole issue of whether there's consciousness in babies and over of what kind, and she certainly thinks that newborn babies, um you know, come into the world with some degree of consciousness. Of course, then you can just extend the question backwards to fetuses and suddenly you're into politically controversial exactly territory. But, you know, they're the question also arises in the animal kingdom. You know, what, where does consciousness start or stop? Is there a line in the animal kingdom where, you know, the first conscious organisms are, it's interesting over time. people are becoming more and more liberal about describing consciousness to animals. People used to think maybe only mammals could be conscious now, most people seem to think show a fish are conscious, they can feel pain. And now we're arguing over insects, you'll find people out there who say plants have some degree of consciousness. So, you know, who knows where it's going to end. The far end of this changes the view that every physical system has some degree of consciousness. Philosophers call that panpsychism. I take that view. </s>

<s>[INST] That you can you can use for good. And tell me your story and spend your really valuable time with me today. This is amazing. It was a great pleasure for me, Lex. Thank you so much. Thanks for listening to this conversation with Brian Keating. To support this podcast, please check out our sponsors in the description. And now let me leave you with some words from Galileo Galilei and questions of science. The authority of a thousand is not worth the humble reasoning of a single individual. Thank you for listening and hope to see you next time. [/INST] </s>

<s>[INST] This is the LaxVietman podcast to support it. Please check out our sponsors in the description. And now to your friends, here's Andrew Strominger. You are part of the Harvard Black Hole Initiative, which has theoretical physicists, experimentalists, and even philosophers. So let me ask the big question, what is a black hole from a theoretical, from an experimental, [/INST] maybe even from a philosophical perspective? So a black hole is defined theoretically as a region of space time from which light can never escape, therefore it's black. Now, that's just the starting point. Many weird things follow from that basic definition, but that is the basic definition. What is light that can't escape from a black hole? Well, light is the stuff that comes out of the sun, that stuff that goes into your eyes. Light is one of the stuff that disappears when the lights go off. This is stuff that appears when the lights come on. Of course I could give you a mathematical definition or physical mathematical definition, but I think it's something that we all understand very intuitively what is light. Black holes on the other hand, we don't understand intuitively, they're very weird. And one of the questions is about black holes, which I think you were alluding to, is why doesn't light get out? Or how is it that there can be a region of space-time from which light can't escape? It definitely happens. We've seen those regions. We have spectacular pictures, especially in the last several years of those regions. They're there. In fact, they're up in the sky, thousands or millions of them. We don't yet know how many. But the proper explanation of why light doesn't escape from a black hole is still a matter of some debate. And one explanation which perhaps Einstein might have given is that light carries energy. You know it carries energy because we have photo cells and we can take the light from the sun and collect it, turn it into electricity. So there's energy in light. And anything that carries energy is subject to a gravitational pull. Gravity will pull at anything with energy. Now, it turns out that there's a lot of energy Now, it turns out that the gravitational pull exerted by an object is proportional to its mass. And so if you get enough mass in a small enough region, you can prevent light from escaping. And let me flesh that out a little more. If you're on the Earth and you're on a rocket ship leaving the surface of the Earth, and if we ignore the friction from the air, if your rocket accelerates up to 11 kilometers per second, that's escape velocity. And if there were no friction, it could just continue forever to the next galaxy. On the moon, which has less mass, it's only seven kilometers per second. But going in the other direction, if you have enough mass in one place, the escape velocity can become the speed of light. If you shine light straight up away from the Earth, it doesn't have too much trouble. It's going way above the escape velocity. But if you have enough mass there, even light can't escape the escape velocity. And according to Einstein's theory of relativity, there is an absolute speed limit in the universe, the speed of light, and nothing makes any sense. Nothing could be self-consistent if there were objects that could exceed light speed. And so in these very, very massive regions of space-time, </s><s>[INST] even light cannot escape. And the interesting thing is Einstein himself didn't think that these objects that would call the black holes could exist. But let me actually linger on this. Yeah, that's incredibly interesting. There's a lot of interesting things here. [/INST] First there's the speed limit. Yeah, that's incredibly interesting. </s><s>[INST] First how wild is it to you if you put yourself in the mind in the time of Einstein before him to come up with the speed limit of, that there is a speed limit? And the speed limit is the speed of light. How difficult of an idea is that? You know, you said from a mathematical physics perspective, everything just kind of falls into place, but he wasn't perhaps maybe initially had the luxury to think mathematically, he had to come up with it intuitively, yes? So like what, how common intuitive is this notion to you? [/INST] Is it still crazy? No, no, so it's a very funny thing in physics. The best discoveries seem completely obvious in retrospect. Even my own discoveries, which of course are far lesser than Einstein's, but many of my papers, many of my collaborators get all confused. We'll try to understand something. We say, we've got to solve this problem. We'll get all confused. Finally, we'll solve it. We'll get it all together. And then we'll, all of a sudden everything will fall into place, we'll explain it. And then we'll look back at our discussions for the proceedings of months and literally be unable to reconstruct how confused we were and how we could ever have thought of it any other way. So not only can I not fathom how confused Einstein was before he, when, you know, when he started thinking about the issues, I can't even reconstruct my own confusion from two weeks ago, you know. So the really beautiful ideas in physics have this, very hard to get yourself back into the mindset. Of course, Einstein was confused about many, many things. Doesn't matter if you're a physicist. It's not how many things you got wrong. It's not the ratio of how many you got wrong, how many you got right. It's the number that you got right. So Einstein didn't believe black holes existed even though he predicted them. And I went and I read that paper, which he wrote, you know, Einstein wrote down his field equations in 1915 and Schwarzschild solved them and discovered the black hole solution three or four months later and very early 1916. And 25 years later, Einstein wrote a paper. So with 25 years to think about what the solution means, wrote a paper in which he said that black holes didn't exist means. And I'm like, well, you know, if one of my students in my general relativity course </s>

<s>[INST] totally. And one other thing from my perspective, because I'm now starting a company. I think the interesting thing about Ellen as he serves as a beacon of hope, Like pragmatically speaking for people that sort of to push back on our doom conversation from earlier that a single individual could build something that allows us as self interested individuals to gather together in a collective way to actually alleviate some of the dangers that face our world. So like it gives me hope as an individual, that I could build something that can actually have impact that counter acts the the Stalin's and the Hitlers and all the threats that face uh that human civilization faces, that an individual has that power. I didn't believe that the individual has that power in the, in the halls of government. Like I don't feel like anyone presidential candidate can rise up and help the world unite the world. It feels like from everything I've seen and you're right with Tesla, it can bring the world together to do good. That's a really powerful mechanism of, you know, whatever you say about capitalism, that you can build companies that start, you know, it starts with a single individual. Of course, there's a collective that that grows around that, but the leadership of a single individual, their ideas, their dreams, their vision can catalyze something that takes over the world and does good for the entire world. [/INST] But if I think, but again, I think the genius of the idea is that it doesn't require us to go head to head with human nature, right? He's he's actually built human nature into the idea by basically saying, I'm not asking you to be, uh, an environmental activist, I'm not asking you to sacrifice to make it, I'm going to sell you a car, you're going to like better and by buying it, you'll help the environment that takes into account our foibles as a species and actually leverages that to work for the greater good. And that's the sort of thing that does turn off my little doom casters cynicism thing a little bit because you're actually hitting us where we live, right? You're you're you're not you can take somebody who doesn't even believe the environments a problem, but they want a Tesla. So they're inadvertently helping anyway. I think that's the genius of the idea. </s><s>[INST] Yeah. And I'm telling you that's one way to make love. Much more efficient mechanism of change than hey, [/INST] making it in yourself interesting to </s><s>[INST] make you suggest creating a product that, uh, at least two more love than hate. [/INST] You're gonna want to love your neighbor because you're gonna make a fortune. I'd get there. That's why I'm on </s><s>[INST] board. That's why yuan said love is the answer. That's I think, uh, exactly what he meant. Okay, let's try something difficult. You've recorded episode steering into the iceberg on your common sense program that has started a lot of conversations. It's quite moving. It was quite haunting. I [/INST] made a lot of angry emails </s><s>[INST] really. Of course [/INST] I did something I haven't done in 30 years. I endorsed a political candidate from one of the two main parties and there were a lot of disillusioned people because of that. </s><s>[INST] I guess I didn't hear it as an endorsement. I just heard it as a, the similar flavor of conversations you have in, in hardcore history. It's almost a speaking about modern times in the same voice as you speak about when you talk about history. So it was just a a little bit of a haunting view of the world today. I know we were just wearing our Doom [/INST] Doom Master put that right back on our, you know, </s><s>[INST] the, uh, I like the term doom castor is there, is there? How do we get love to win? What's the way out of this? Is there some hopeful line that we can walk to a to avoid something? And I hate to use the terminology, but something that looks like a civil war, not necessarily a war of force, but a division to a level where it doesn't any longer feel like a United States of America with the emphasis on United. Is there a way out? [/INST] I read a book a while back, I want to say George Friedman. The street for a guy wrote it was something called The Next 100 Years, I think it was called. And I remember thinking um I didn't agree with any of it. And one of the things I think he said in the book was that the United States was going to break up. I'm going from memory here. He might not have said that at all, but something was stuck in my memory about that. I remember thinking, um, but I think some of the arguments were connected to the differences that we had and the fact that those differences are being exploited. So we talked about media earlier in the lack of truth and we have a media climate that is incentivized to take the wedges in our society and make them wider and there's no countervailing force to do the opposite or to help, you know? So there was a famous uh memo from a group called project for a new american century and they took it down. But the way back machine online still has it. And it happened before 9 11 spawned all kind of conspiracy theories because it was saying something to the effect of uh and I'm really paraphrasing here. But you know that the United States needs another Pearl Harbor type event because those galvanized a country that without those kinds of events periodically is naturally geared towards pulling itself apart. And it's those periodic events that act as the countervailing force that otherwise is not there. Um, if that's true, then we are naturally inclined towards pulling ourselves apart. So to have a media environment that makes money off widening those divisions, which we do. I mean, I was in talk radio and and it has those people, the people that used to scream at me because I wouldn't do it. But I mean, we would have these terrible conversations after every broadcast where I'd be in there with the program director and they're yelling at me about heat. Heat was the worthy create more heat. Well, what is heat right here? This division? Right? And they want the heat, not because their political, they're not republicans or democrats either there. We want listeners and we want engagement and involvement. And because of the constructs of the format, you don't have a lot of time to get it. So you can't have me giving you like on a podcast an hour and a half or two hours where we build a logical argument And you're with me the whole way your audience is changing every 15 minutes. So whatever points you make to create interest and intrigue and engagement have to be knee jerk right now. Things they told me once that the audience has to know where you stand on every single issue within five minutes of turning on your show. In other words, you have to be part of a of a linear set of political beliefs. So that if you feel a about subject a then you must feel de about subject D. And I don't even need to hear your opinion on it, because if you feel that way about a you're going to feel that way about D. This is a system that is designed to pull us apart for profit, but not because they want to pull us apart, right? It's a byproduct of the Prophet. That's one little example of a 50 examples in our society that work in that same fashion. So what that project for a new American century document was saying is that we're naturally inclined towards disunity and without things to occasionally ratchet the unity back up again. So that we can start from the baseline again and then pull ourselves apart till the next Pearl Harbor that you'll pull yourself apart, which I think was I think that's what the George Friedman book was saying that I disagreed with so much at the time. Um, so in answer to your question about civil wars, we can't have the same kind of civil war because we don't have a geographical division. That's as clear cut as the one we had before. Right? You have to basically north south line and some border states, it was set up for that kind of split. Now we're divided within communities, within families, within gerrymandered voting districts and precincts, right? So you can't disengage we're stuck with each other. So there's a civil war now, for lack of a better word, what it might seem like is the late 19 sixties, early 19 seventies where you had the bombings and you know, let's call it domestic terrorism and things like that because that that would seem to be something that once again, you don't even need a large chunk of the country. Pulling apart 10% of people who think it's the end. Times can do the damage. Just like we talked about terrorism before and a can of gas and a big lighter. I've lived in a bunch of places and I won't give anybody ideas where a can of gas and a big lighter would take 1000 houses down before. You could blink right? Um, that terrorist doesn't have to be from the Middle East doesn't have to have some sort of a fundamentalist religious agenda. It could just be somebody really piste off about the election results. So once again, if we're playing an odds game here, everybody has to behave for this to work right. Only a few people have to misbehave for this thing to go sideways and remember for every action there's an equal and opposite reaction. So you don't even have to have those people doing all these things, All they have to do is start retribution cycle. </s>

<s>[INST] that? It's incredible! So you can have this fingerprint within the space of colors and there's a lot of molecules. And I mean I I wonder if it's a question of how much overlap there is how how close can you get to the actual Fingerprint like can foster and unlock the iPhone with its lights on. He says 16.8 billion. So presumably this rainbow is dis critized into little segments somehow. How many total are there? How How How a Lot is 16.8 billion. [/INST] It's a lot. You we don't have the instruments to break these, break any light into this many tiny segments. And so with the instruments we do have, there's huge amounts of overlap, methane as an example. A lot of the ways it's um detectable is because the carbon and the hydrogen is they vibrate with one another. They move, they interact but every other hydrocarbon are settling. IsA prin has carbon and hydrogen also vibrating and rotating. And so it's actually very hard to tell them apart at low resolutions. And our instruments can't really cope with distinguishing between molecules particularly well. Yeah. But in an ideal world, if we had infinite resolution then yes every molecule spectral features will be unique. </s><s>[INST] Yeah. Like almost like it would be too trivial [/INST] level at the quantum level at our level there's huge overlap. </s><s>[INST] But then you can start to then what tried to dis a big you ate like what the miss the fact that certain colors are missing. What does that mean? And hopefully they're missing a certain kind of pattern where you can say with some kind of probability that this gas not this gas. You're solving that gashes puzzle. I got it. We [/INST] can go back to venus actually and show that so with this I mentioned those two molecules that could be responsible for that signal. The resolution that we have. It was fasting and S. O. To um sulfur dioxide and at that resolution could really be one of the other. But in the same bandwidth. So in the kind of the same observations there was another region where fossil does not absorb, we know that but so too does so we just went and checked and there was no signal. So I thought oh then it must be fascinating. And then we submit to the paper. The rest is history. </s><s>[INST] Uh Well yeah that's um beautifully told is there. So the telescopes we're talking about are sitting on earth. What can it help solving this fingerprint? Molecular fingerprint problem If we do a fly by, does it help if you get closer and closer or our telescopes pretty damn good for this kind of puzzle solving [/INST] skills are pretty good. But the earth's atmosphere is a pain. I mean I'm very thankful for it but it does interrupt a lot of measurements and a lot of regions where fossil would be active, they are not available. The earth is not transparent in those wavelengths. So being above the atmosphere would make a huge difference then proximity matters a lot less but just escaping the earth's atmosphere would be wonderful. But then it's really hard to stay very stable and if there is fasting on venus there is very little of it in the clouds. And so the signal is very weak and the telescopes we can use on earth are much bigger and much more stable. So it's a bit of a trade off. </s><s>[INST] So is it uh are you comfortable with this kind of remote observation? Is it at all helpful to strive for going over to venice and like grabbing a scoop of the atmosphere or his remote observation? Really a powerful tool for this kind of job like the scoop is not necessary. [/INST] Well a lot of people want to scoop. I get it. I get it completely. I I don't want to scoop specifically because if it is life, I want to know everything I can remotely before I interfere. Um So that's my, I've got ethical reasons against the scoop more than engineering reasons against the scoop. But I have some engineering reasons against the scoop. Scoop is not a technical term, but it's too </s><s>[INST] late going along [/INST] with this. It's too late to take it back. I appreciate it. We don't understand the clouds well enough to plan this scoop very well </s><s>[INST] because it's not that saturated, like there's not that much of it present. [/INST] No. And the places nasty, you know, it's not, it's not going to be easy to build something that can do the task reliably and can be trusted. The measurements can be trusted and then pass that message on. So actually I'm for an orbiter, I think we should have orbiters around every solar system body whose job is just to learn about these places. I'm disappointed. We haven't already got an orbited around every single one of them. A small, it can be a small satellite exploding data, figuring out, you know, how do the clouds move? What's in them? How often is there lightning and volcanic activity? Whereas the topography, is it changing? Is there a biosphere actively doing things. We should be monitoring this from afar. And so I'm for a over the atmosphere, hopefully around venus, that would be that would be my choice. </s><s>[INST] Okay, so now recently venus is all exciting about phosphene and everything? Is there, is there other stuff maybe before we were looking at venus or now looking out into other solar systems? Is there other promising exoplanets or other planets within the solar system that might have phosphene or might have other strong biosignatures that we should be looking for. Like phosphene, [/INST] there's a few, but outside the solar system all are promising candidates. We know so little about them. For most of them, we barely know their density. Um most of them we don't even know if they have an atmosphere, never mind what that atmosphere might contain. So we're still very much at the stage where we have detected promising planets, but they're promising in that they're about the right size, about the right density. They could have an atmosphere and they're about the right distance from their host star. But that's really all we know near future telescopes will tell us much more. But for now we're just guessing </s>

<s>[INST] Do you have other worries, like some people like Elon and others have worries of existential threats of artificial intelligence of artificial general intelligence or if you think more narrowly about threats and concerns about more narrow artificial intelligence, like what are your thoughts in this domain? You have concerns are you more optimistic? I think [/INST] You've all in his in his book 21 lessons for the 21st century, his last chapters around meditation. And uh you look at the title of the chapter and you're like, oh it's kind of you know, it's all meditation. But the what was interesting about that chapter is he believes that um you know, kids being born today growing up today, um Google has a stronger sense of their preferences than they do, which you can easily imagine. I can easily imagine today that google probably knows my preferences more than my mother does. Um maybe not me per se, but for someone growing up only knowing the internet only knowing what google is capable of or facebook or twitter or square or any of these things. Um the self awareness is being offloaded to other systems and particularly these these algorithms. And his concern is that we lose that self awareness because the self awareness is now outside of us and it's doing such a better job at helping us direct our decisions around. Should I stand. Should I walk today? Um what doctor should I choose? Who should I date all these things? We're now seeing play out very quickly. So he sees meditation as a tool to build that self awareness and to bring the focus back on why do I make these decisions? Why do I react in this way? Why did I have this thought? Where did that come from? </s><s>[INST] That's the way to regain control. [/INST] Your awareness may be not control but but awareness so that you can be aware that yes, I am offloading this decision to this algorithm that I don't fully understand and can't tell me why it's doing the things that's doing because it's so complex. </s><s>[INST] That's not to say that the algorithm can't be a good thing. And to me recommend our systems. The best of what they can do is to help guide you on a journey of learning new ideas of of learning period. [/INST] It can be a great thing. But do you know you're doing that? Are you aware that you're inviting it to do that to you? I think that's that's the risk he identifies, right? Um is uh that's perfectly okay. But are you aware of that? Do you have that invitation? And it's it's being acted upon. </s><s>[INST] And so that that's your that's a concern. You're kind of highlighting that without a lack of awareness, you can just be like floating at sea. So, awareness is key in in the the future of these artificial intelligence systems. [/INST] The other movie wall e which I think is one of Pixar's best movies besides predatory credit to you. Was you </s><s>[INST] had me until right into [/INST] it. Okay. The tour is </s><s>[INST] incredible. Mm hmm. All right. We've come to the first point where we disagree. Okay, [/INST] see entrepreneurial story in the form of Iraq. </s><s>[INST] Mhm. I just remember just the soundtrack was really good. So excellent. What are your thoughts sticking on artificial intelligence? A little bit about the displacement of jobs? That's another perspective that candidates like Andrew [/INST] yang talk about yang gang </s><s>[INST] forever, yang gang. So he unfortunately speaking of yang gang is recently dropped [/INST] out. It was very disappointing and depressing. </s><s>[INST] Yeah, but on the positive side, he's I think launching a podcast. So uh really cool. Yeah, he just announced that I'm sure he'll try to talk you into trying to come on to the podcast. [/INST] So about predatory. </s><s>[INST] Yeah, maybe he'll be more welcoming of the repertory argument. What are your thoughts on his concerns that the displacement of jobs of automation is that of the of course there's positive impacts that could come from automation and ai but it could also be negative impacts. And within that framework, what are your thoughts about universal basic income? So, these interesting new ideas of how we can empower people in the economy. [/INST] I I think he was 100% right on almost every dimension. Um we see this in squares business. I mean he identified um truck drivers, I'm from Missouri. Um and he certainly pointed to the concern and the issue that people from where I'm from, feel every single day that is often invisible and not talked about enough. The next big one is cashiers, this is where it pertains to Square's business. Um we are seeing more and more of the the point of sale moved to the individual customer's hand in the form of their phone and apps and pre order and order ahead. We're seeing more kiosks, we're seeing more things like amazon go and the number of workers in as a as a cashier and re tos immense and you know there's no real answers on how they transform their skills and work into into something else. And I think that does lead to a lot of really negative ramifications and the important point that he brought up around universal basic income is given that the shift is going to come and given it is going to take time two um set people up with new skills and new careers, they need to have a floor to be able to survive and This $1,000 a month is such a floor um it's not going to incentivize you to quit your job because it's not enough but it will enable you to not have to worry as much about just getting on day to day so that you can focus on what I'm what am I going to do now and what am I going to um what skills do I need to acquire. Um and I think I think you know a lot of people point to the fact that you know during the industrial age we we have the same concerns around automation factory lines and everything worked out okay. But the biggest change is just the the velocity and the centralization of a lot of the things that make this work which is the data um and the algorithms that work on this on this data. I think the second biggest scary thing is just how around ai is just who actually owns the data and who can operate on it. And um are we able to share the insights from the data so that we can also build algorithms that help our needs our help our business or whatnot? So that's where I think regulation could play a strong and positive part. Um First looking at the primitives of AI and the tools we use to build these services that will ultimately touch every single aspect of the human experience and then how data um where data is owned um and how it's how it's shared. Um So those those are the answers that as a society as a world we need to have better answers around which we're currently not, they're just way too centralist into a few very very large companies but I think it was spot on with identifying the problem and proposing solutions that would actually work at least that we learned from that you could expand or evolve. Um But I mean it's I think it's U. B. I. S. Well well past its its do I mean it was certainly trumpeted by martin Luther king and even even before him as well </s>

<s>[INST] useless. And there's a gray area, maybe you can comment on it, is sometimes a drop of compassion helps make that empathy more effective in the conversation. So you're just saying you believe [/INST] x doesn't quite form a strong of a bond with the other person, the other person. </s><s>[INST] You're imagining it doesn't. Maybe you're right, yes, I'm imagining it doesn't. I'm imagining you need to show that you're on the same side. You need to signal a little bit about your actual beliefs, at least in that moment, even if that signaling is not as deep as it sounds. But at first, basically patting the person on the back and saying, we're on the same side, brother. [/INST] You know, that's what most people, when they're really learning the concept, that's the basic human reaction. And in application, especially in highly adversarial situations, like, I need a regular guy, Muslim, but how's that guy going to say, buy it if I like, you know, dude, I'm on your side. I've been there. I feel you. No, no, no, no, no, no. People get conned by that so much. Like if we're on opposite sides of the table, and I try to act like I'm not on the opposite side of the table, that makes me disingenuous. So I would rather be honest. My currency is integrity. And at some point in time, if you go like, you know where I'm coming from? My answer is going to be like, look, I can agree on maybe where we're going. But if we're talking about, you know, am I on your side now? As a human being, I want to see you survive and thrive, not at my expense. I think the world is full of opportunity. I'm optimistic. And I got more than enough reason for saying that. It's enough for here for both of us. So I got no problem with you getting yours. You know, just don't take it out of my hide. And I'm going to be honest about it, about both of those things. I'm not interested in you taking it out of my hide. I think there's plenty here for both of us. Now, I don't need to be on your side, except in a human sense. But do I have to side with you over the war? No. Or how we're distributing the stock, or how much you get paid, or how much you make off this car. I think people, my experience as a layman, is that empathy is not got a downside. That you don't need me to act like </s><s>[INST] I'm on your side for us to make a great deal. Great deal. Well, we'll talk about two things, a great deal and a great conversation. They're often going to be the same thing, but at times they're going to be different. You mentioned Vladimir Putin. There is some zoom level at which you do want to say we're on the same side. You said the human level. It's possible to say, kind of zoom out and say that we're all in this together. Not we Slavic people, we Europeans, [/INST] but we human beings. Well, the same planet. Same planet. Right. Several years ago, and his name has evidently been mud now, but he was very nice to me. A lawyer here in a town named Tom Girardi, and no shortage of bad reporting on him now. I have absolutely no idea if any of it's true. I do know that in my interaction with him, he was always a gentleman to me and was very generous. When he'd get into conversations with people, he'd always say, let's look at 10 years from now, where we could both be in a phenomenal place together. Now let's work our way back from there. That's a good line. Yeah. And then I saw him do it in simulations. I was teaching at USC. We were at a function together and a gentleman at the time told me that who he was and he was really influential. So I walked up to the guy, Cold that I said, Hey, I'm going to come into talking to my class, he didn't know me. I learned the fact that </s><s>[INST] we had mutual acquaintance and he graciously consented to come in. [/INST] Yeah. And he said, what do you want me to talk about? And I said, look, dude, just from your success here, it doesn't matter what you talk about, either I'm going to agree or I'm going to disagree, or I'm going to learn from it, my students are going to learn from it. So students want to role play with them. They dispute and let's do a negotiation. Every single time you'd go to pick a point in the future where we're both happy 10 years, 20 years from now, and let's work our way back. Now, hostage negotiator, same thing. I call into a bank, bad guy picks up on the phone, and I'm going to say, I want you to live. I want to see you survive this. Whatever else goes with that, let's pick a point in the future that we're both good with, and then we work our way back. And people make also, we were talking before about emotion and what you care about, people make their decisions based on their vision of the future. Without question, I think there's a Hindu temple in the United States has been or being assembled the same way that the Hindu temples were in India a thousand years ago, by hand, volunteers, by hand. These people are knocking themselves on for place in paradise, a vision of the future. What you will go through today, if the future portends what you want, you'll go through incredible things today. It's a vision of the </s><s>[INST] future. So you have to try to paint a vision of the future that the person you're negotiating [/INST] with will like. Just top to do. Yeah. Let's find out what their vision of the future is, is, and then remove yourself as a threat. If we can collaborate together at all, if you think that I could do anything at all to help you to that point, and integrity is my currency, I'm not going to lie to you, which gets back before that I lied to you about whether or not I'm on your side. Right now, at the moment, we're on opposite sides of the fence. That's not going to stop us from being together in the future. You're going to say, well, you didn't lie to me about today, maybe you won't lie to </s>

<s>[INST] Um Would you say that's if we're looking at the future of human robot interaction? Would you say the psychology piece is the hardest, like if it's I mean it's a funny notion for you as I don't know if you consider. Yeah I mean one way to ask it, do you consider yourself a roboticist or psychologist? [/INST] Oh I consider myself a robotics. Is that plays the act </s><s>[INST] of a psychologist. But if you look at yourself sort of um you know 2030 years from now, do you see yourself more and more wearing the psychology hat? Uh sort of another way to put it is are the hard problems in human robot interactions fundamentally psychology or is it still robotics? The perception manipulation planning and all that kind of stuff. [/INST] It's actually neither the hardest part is the adaptation in the interaction. So it's the interface is the learning. And so if I think of like I've become much more of a roboticist slash ai person than when I originally again I was about the biotics, I was I was electrical engineer, I was control theory right? Like and then I started realizing that my algorithms needed like human data right? And so then I was like okay what is this human thing? Right. How do I incorporate human data? And then I realized that human perception had like there was a lot in terms of how we perceive the world and still trying to figure out how do I model human perception from my and so I became a HR I person human robot interaction person from being a control theory and realizing that humans actually offered quite a bit. Uh and then when you do that you become more of an artificial intelligence ai And so I see myself evolving more in this ai world um under the lens of robotics having hardware interacting with </s><s>[INST] people. So you're a world class expert researcher and robotics and yet others, you know there's a few is it's a small but fierce community of people but most of them don't take the journey into the age of HR I into the human. So why did you brave into the interaction with humans? It seems like a really hard problem. [/INST] It's a hard problem. It is very risky as an academic. And I knew that when I started down that journey that it was very risky as an academic um in this world. That was nuance. It was just developing. We didn't even have a conference right at the time because it was the interesting problems. That was what drove me. It was the fact that I looked at what interests me in terms of the application space and the problems and that pushed me into trying to figure out what people were and what humans were and how to adapt to them. Um If those problems weren't so interesting, I'd probably still be sending rovers to glaciers, right? Uh But the problems were interesting and the other thing was that they were hard, right? So it's I like having to go into a room and being like I don't know what to do and then going back and saying okay, I'm going to figure this out. I do not, I'm not driven when I go and like oh there are no surprises, like I don't find that satisfying. Um If that was the case I go someplace and make a lot more money right? I think I stay in academic because and choose to do this because I can go into a room like that's hard. </s><s>[INST] Yeah, I think just for my perspective maybe you can correct me on it? But if I just look at the field of ai broadly it seems that human robot interaction has the most one of the most uh number of open problems like people, especially relative to how many people are willing to acknowledge that there are. This is because most people are just afraid of the humans so they don't even acknowledge how many open problems there. But it's a in terms of difficult problems to solve exciting spaces, it seems to be incredible for that. [/INST] It is it is exciting. </s><s>[INST] You mentioned trust before. What roll the trust from interacting with autopilot to in the medical context. What role does trust playing the human robot [/INST] interaction? So some of the things I study in this domain is not just trust, but it really is over </s><s>[INST] trust. How do you think about over trust? Like what is first of all, what is uh what is trust and what is over trust [/INST] basically the way I look at it is trust is not what you click on a survey trust, this is about your behavior. So if you interact with the technology based on the decision are the actions of the technology as if you trust that decision, then you're trusting right. Um And even in my group we've done surveys that you know on the thing, Do you trust robots? Of course not. Would you follow this robot in a burning building? Of course not. Right. And then you look at their actions and you're like clearly your behavior does not match what you think right or what you think you would like to think right. And so I'm really concerned about the behavior because that's really at the end of the day when you're in the world, that's what will impact others around you is not, whether before you went onto the street you clicked on. Like I don't trust self </s>

<s>[INST] I felt could have made his case a little bit stronger. So the way, the thing he succeeded is he didn't come off as like a science expert looking down at everybody, talking down to everybody. So he succeeded in that, which is very respectful. But I also think sort of making the case for taking the vaccine when you're a young, healthy person, when you're sitting across from Joe Rogan is like a high difficulty on the video game level. For sure. So it's not, it's difficult to do. Yeah, for sure. It's difficult to do. And also it's difficult to do because it's not as simple as like, look at the data. There's a lot of data to go through here. And there's also a lot of non-data stuff, like the fact that, first of all, questioning the sources of the data, the quality of the data. Cause it's also disappointing about COVID is that the quality of the data is not great, but also questioning all the motivations of the different parties involved, whether it's major organizations that developed the vaccine, whether it's major institutions like NIH or NIAID that are sort of communicating to us about the vaccine. whether it's the CDC and the WHO, whether it's the Biden or the Trump administration, whether it's China and all those kinds of things, you have to, that's part of the conversation here. I mean, vaccination is not just a public health tool, it's also a tool for a government to gain more control over the populace. Like there's a lot of truth to that too. Things that have a lot of benefit can also be used as a Trojan horse to increase bureaucracy and control. [/INST] But that has to be on the table for a conversation. Yeah, I think it has to be on the conversation. But I mean, your parents, when they were in the Soviet Union and here in the United States, and actually it was a big collaboration between US and Soviet Union, when the polio vaccine came out, there were people all around the world who had a different life trajectory, no longer living in fear. And all of these people who were paralyzed or killed from polio, smallpox has been eradicated. It was one of the great successes in human history. And while it for sure is true, you could imagine some kind of fraudulent vaccination effort. But here, I genuinely think, I mean, whatever the number, 15 million, 16 million is the economist number of dead from COVID-19, many, many, many more people would be dead but for these vaccines. And so I get that any activity that needs to be coordinated by a central government has the potential to increase bureaucracy and increase control. But there are certain things that central governments do like the development, particularly these mRNA vaccines, which it's purely a US government victory. I mean, there was huge DARPA funding and then the National Institute for Allergy and Infectious Disease, NIH funding. I mean, this was a public private partnership throughout and that we got a working vaccine in 11 months was a miracle. </s><s>[INST] So it's not purely a victory. Again, you have to be open-minded. I'm with you here playing a bit of devil's advocate but the people who discuss antiviral drugs like hypermectin and other alternatives would say that the extreme focus on the vaccine distracted us from considering other possibilities. And saying that this is purely a success is distracting from the story that there could have been other solutions. So yes, it's a huge success that the vaccine was developed so quickly and surprisingly way more effective than it was hoped for but there could have been other solutions and they completely distracted from that. In fact, it distracted us from looking into a bunch of things like the lab leak. And so it's not a pure victory. And there's a lot of people that criticize the overreach of government and all of this, that one of the things that makes the United States great is the individualism and the hesitancy to ideas of mandates. Even if the mandates on mass will have a positive, even strongly positive result, the Americans, many Americans will still say no because in the long arc of history, saying no in that moment will actually lead to a better country and a better world. So that's a messed up aspect of America but it's also a beautiful part. [/INST] We're skeptical even about good things. I agree and certainly we should all be cautious about government overreach, absolutely. And it happens in all kinds of scenarios with incarceration, with a thousand things. And we also should be afraid of government underreach that if there is a problem that could be solved by governments and that's why we have governments in the first place is that there's just certain things that individuals can't do on their own and that's why we pool our resources and we in some ways sacrifice our rights for this common thing. And that's why we don't have hopefully people, murderers marauding or people driving 200 miles down the street that we have a process for arriving at a set of common rules. And so while I fully agree that we need to respect and we need to listen, we need to find that right balance and you've raised the magic I word, ivermectin. And so an ivermectin, like my view has always been ivermectin could be effective, it could not be effective. Let's study it through a full process. And when you had Francis Collins with you, even while he was making up stories about this wrestler, he was saying, yeah, exactly. But he was saying that they're going to do a full randomized highest level trial of ivermectin. And if ivermectin works, then that's another tool in our toolbox. And I think we should. And I think that Sanjay was absolutely correct to concede the point to Joe that it was disingenuous for people, including people on CNN to say that ivermectin is for livestock. And so I definitely think that we have to, like we have to have some kind of process that allows us to come together. And I totally agree that the great strength of America is that we empower individuals. It's the history of our frontier mentality in our country. So we, I 100% agree that we have to allow that, even if sometimes it creates messy processes </s>

<s>[INST] And then you came back to uh I mean, this is the hard thing after this heartbreak to step up and continue fighting, right? [/INST] I really, really struggled, Like, unbelievably struggled from 2012 to 2014. I almost quit numerous times. I was so angry. I mean, at one I got so mad at the idea of feeling like they were fucking me every step of the way. I threw a water bottle at the referee after a match, I cussed out a referee one time on a mat, I got suspended from the sport because I was just so angry at that point in time. </s><s>[INST] And I G. F. Is the International judo federation. And there are there the people that supply the referees, basically like the kind of [/INST] run the sport that on a global scale. </s><s>[INST] So uh you sent a few emails 2014, 15, uh basically quitting one of them, said I'm mentally and physically broken. Another said, well, the subject line, I'm done, yep, the weight cuts didn't break, you know? So if this broke you, you were really going through a hard time. [/INST] I was like, you know what? We're just gonna like, dumb it down a little bit and get some wins under our belt. I'm going to go to a World cup, which is like three stages down our four stages down from like the olympic games. Like this should be like a cakewalk, like making the final level World Cup should be a walk in the park. I show up, I barely beat a 16 year old kid. Yeah, barely. Then I got smoked in the second round, I got thrown three times. I was like, I'm fucking done. They changed all the fucking rules, they fucked me out of the olympics, like, what am I supposed to do? And it was at that moment when I wrote the email where I remember sitting at a bar, I don't drink by the way, but I was sitting at a bar um at the hotel sending this email and I got a response back from jimmy and he goes, well, just just stay for the training camp, go to Germany and then whatever happens, don't worry about it. We'll talk when you get home. I was like, fuck that, fuck these people, fuck the rules, I don't fucking care anymore. I'm just gonna do judo the way I want to do judo if I fucking get she towed out. Fuck them. That was my response. </s><s>[INST] Can you become an olympic champion? Can you become an olympic medalist with that kind of thinking you think or know, was that that's counterproductive. Okay, just, just checking because maybe that's also liberating. [/INST] The expectation was no longer that Travis is going to win this tournament. The expectation was Travis is going to come home and be fucking pissed off. We're gonna have to figure out how to manage a pissed off person that's trying to quit, That shouldn't be quitting. </s><s>[INST] And the people still believe that you can be a medalist again. Like, who believed [/INST] that jimmy believed it, the team managers believed it. Some of my teammates still believed it. Um My training partner still did, but they're not the ones that are cutting the weight, flying around, feeling like, you know, all of your judo is now null and void, right? Because at this point they took away leg grabs entirely, you couldn't break a grip with two hands, right? The meta of judo has changed again, right? So I got fucked out of it, they took away how I did judo again, and now it just got more difficult. So when I'm sitting in the hotel and I'm sending this email, I remember being at the training camp like, I was like, I don't even fucking character rules are I'm just gonna fucking throw people, I don't even care from cheating, it doesn't matter to me, I'll just play stupid. Yeah, right, So I just started going back and doing judo without the leg grabs, but with all the same gripping that I was doing beforehand. And then when I got to Germany I was like, I don't fucking care, I was like, if I got to cheat to win, then I got a fucking cheat to win. If I get cheated out? Like, then I get cheated out, and I want Germany </s><s>[INST] uh would you want to Germany [/INST] the german grand prix, which is a a week after losing the World Cup because I was trying to do do judo around the new rule set, I wasn't just trying to do judo, right? Because when you get to the highest level, your game tends to morph around, you know what can you can or cannot get away with. I was more focused on trying to figure out what I can and can't get away with. And I stopped actively doing judo once I said, fuck! Whatever the rule changes are, I'm just going to keep doing judo the way I know how to do judo. And if I get a penalty then soviet. </s><s>[INST] And so that that when that started the road [/INST] back Yeah, because now it's like, I don't care if you penalize me or not because I'm gonna throw that guy anyways, I'm gonna beat them anyways, and if I get a shido for doing something wrong, then I'll just stop doing that one thing and just keep doing all the other things that they told me I probably shouldn't be doing, but they're not calling me on it, so I'm just going to keep doing it, </s><s>[INST] wow you uh You find yourself to 2016 Olympics. Was that ever a doubt by the way after this, after 2014 in Germany [/INST] I had a lot of doubt after the concussion in 2015, I remember when I first came back after four months of nothingness. That like even trying to like train the room would start to like tilt the world on me and then when I finally got over that and I could start doing things again, I stepped on the mat for the pan am's and I was like, drowning is not the right word, but like everything was being done in such a slow motion. Like I had sandbags everywhere that like, I just couldn't keep up </s>

<s>[INST] That's great. Yeah. As I mentioned to you offline, I talked to Matthew johnson is a Hopkins professor and uh, he's a really a scholar of most, he's starting most drugs. He's also really deeply studied cocaine, all those stuff, negative effects and uh he's focused on a lot of positive effects of the different psychedelics. It's kind of fascinating. So I'm very much interested in exploring the science of what these things do to the, to the human mind and also personally exploring it, although it's like this weird gray area, which he's masterful at, which is, he's a professor at john Hopkins, one of the most prestigious universities in the world and and doing large scale studies of the stuff and until until he got a lot of money for these studies, even in Hopkins itself, there's not much respect, even respect. It was like people just didn't want to talk about it as a, as a legitimate field of inquiry. It's kind of fascinating how hesitant we are as a little human civilization to legitimize the exploration of the mysterious of whatever the definition of the mysterious is for that particular period of time. So, for us now, there's like, little groups of things, like, I would say consciousness in the space of, like, computer science research is something that's still like, I don't know, maybe less philosophers kicking around for a little longer, and then certainly extraterrestrial life forms from um in most formulations of that problem space uh is still the other, it's still the source of the mysterious except maybe like, cT, which is like, how can we detect signals from far away alien intelligences that will be able to perceive. Yeah, it's uh and psychedelics is another one of those that's like, we're starting to see, okay, well, can we um try to see if there's some medical applications of, like, helping you get, like, he does studies of help you quit smoking or help you in some kind of treatment of some disease? And he's sneaking into that. I mean, it's like, openly sneaking into it. Some studies on it of like, how can you expand the mind with these tools and what can the mind discover through psychedelics and so on. And we're, like, slowly creeping into the space of being able to explore these mysterious questions. But it's like, it sucks that sometimes a lot of people have to die, meaning, sorry, they have to age out. Mhm. Like it's like faculty have uh and people have a fixed set of ideas and they stick by them and in order for new ideas to come in, then the young folks have to be born with with an open mind the possibility of those ideas and then they have to become old enough and get A's in school or whatever to to take care of those ideas forward. So, you know, the acceptance of the exploration of the mysterious takes time. It's kind of sad. [/INST] It is sad. I </s><s>[INST] agree. Maybe to go into my source of passion, which is artificial intelligence. What's your sense about the the possibility? Uh like Pamela McCormick has this quote that I like, I talked to a couple of years ago, I guess already in this podcast that that artificial intelligence began with the answer wish to forge the gods. So do you think artificial intelligence may become the very kind of gods that were at the center of our religions of most of our history? [/INST] Yeah, there's a lot there. So I'm going to start by addressing this idea of artificial intelligence being separate from human beings. Okay, so that I don't think that's actually, that might happen. Okay, I mean it's already happened, but let's put it this way, like you're talking about super artificial intelligence, like autonomous, conscious artificial intelligence. Okay. Yeah. </s><s>[INST] Something with artificial consciousness. [/INST] First of all I think she's correct. Okay, but also it's awesome. An awesome quote. I'd also like to bring up this uh writer of fiction actually. Um Ted Chiang and his one of his essays, he writes short essays. One of them was the basis for the movie arrival which if you haven't seen it, it's a really great movie about uh UFOs and </s><s>[INST] has a very creative way of proposing an idea of how they might be able to communicate first of all how they appear to us. Second of all how they may be communicating with us humans. [/INST] Exactly. And the author Ted Chiang has a lot, I recommend his writings, his short stories. Someone is very short and it appeared as It appeared in nature about 20 years ago and it is called I think it's called getting eating the crumbs from the table or something like that. And it's basically this short essay and I hate to do you know to to do a spoiler here, but if you don't want to know what it's about, don't listen right now. Yeah. Okay so this is what it's about. So basically it's about human beings uh becoming two different species. Okay. And one of them is created, they're called meta humans and they start bio hacking themselves with tech. Okay, sound familiar. So they do this and they become meta humans and another species. Right? And you know just kind of another fork such that humans can barely understand them because they're so far removed. So in a sense are very gods, right? No, they're meta humans are superhumans are enhanced humans. Okay. I see that hopefully on the horizon, frankly, I hope so. Not that we have two species but that we can use our technology or we can we can become so integrated with our technology that we can survive, Okay, we can survive the radiation in space. We can't go places now because of the radiation in space. Perhaps we can develop our bodies such that we can survive the radiation in space. So there's this idea of these meta humans now, there's also this idea that technology is just another form of humans. We've created it. Right? And so maybe it is bent on surviving there by using us, you know, kind of as a meme or a team. Some people are calling them teams now, these self generating uh they're replicating themselves through us. Okay, I see that also and I don't think that's terribly bad. Maybe it's just the way that we are evolving. It doesn't mean that the, you know, we're evolving all the time. Like we're taller than we used to be. You know, we have different skills and you know, so I don't see that as as a bad thing. I think a lot of people see it as if we're not how we are now. It's a tragedy, but it's not a tragedy, how we are now is actually a tragedy for most people alive. </s>

<s>[INST] So freaking cool. That's awesome. So you mentioned microgravity and I saw, I believe that there's a picture of you floating in microgravity. When when did you get to experience that? What was that like? [/INST] So I've flown nine times on the affectionately known as the vomit comet. It's the parabolic flight And essentially it does what you want to play never to do. It pitches really steeply upwards at 45 </s><s>[INST] degrees. [/INST] Yeah, that's test. Right. That's super early in my PhD some of just the passive tiles before we even put electronics. And we were just testing the magnet polarity and the essentially is it an energy favorable structure to self assemble on its own? So we tweaked a lot of things between. </s><s>[INST] Are we looking at a couple of them? [/INST] Yeah. You're looking at a bunch of them. There's </s><s>[INST] almost 30 [/INST] two of them. </s><s>[INST] Yeah, they're clumping [/INST] clumping. Can </s><s>[INST] you comment on what's the difference, jean microgravity and and zero gravity? [/INST] Yes. So there is, it's important difference. There is no zero gravity, there's no nothing, there's in the universe. There is no such thing as zero gravity. So Newton's the law of gravity tells us that there's always gravity attraction between any two objects. So zero G. Is a shorthand that some of us fall into using. It is a little easier to communicate to the public. The accurate term is microgravity. Where you are essentially floating. You're weightless but generally in free fall. So on the parabolic flights, the vomit comet, you're in free fall at the end of the parabola and in orbit around the Earth when you're floating, you're also in free fall. </s><s>[INST] What [/INST] was it like? Uh </s><s>[INST] so affectionately call vomit comment. I'm sure there's a reason why it's called affectionately. So so what's it like? What's uh your first time to both philosophically spiritually and biologically. What's it like? [/INST] It's profound. It is unlike anything else you will experience on earth because it is this true feeling of weightlessness. With no drag. So the closest experience you can think of would be floating in a pool, but you move slowly when you float in a pool and your motion is restricted when you're floating. It's just you and your body flying like in a dream. Um It takes the littlest amount of energy, like a finger tap against the wall of the plane to shoot all the way across the fuselage </s><s>[INST] and you can move at full speed, you can move your arms exactly your muscles. [/INST] There's no yeah, </s><s>[INST] there's no resistance. And they [/INST] actually tell you to make a memory when you're on the plane because it's such a fleeting experience for your body that even a few days later you've already forgotten exactly what it felt like. It's so foreign to the human experience. </s><s>[INST] They kind of suggested you explicitly tried to really form this into a memory that you could do the replay training save when we have neural link, we can replay that that that memory. So in terms of how much stress it has on your body, is it biologically stressful? [/INST] You do feel A to G pullout? Right? So the cost of getting those micro g parabolas is you then have a to G pullout and that's hard, you have to train for it if you move your neck too quickly on that to g pull out, you can strain muscles. But I wouldn't say that it's actually a a profound um tough thing on the body, it's really just an incredibly novel experience and when you're in orbit and you're not having to go through the ups and downs of the parabolic plane, there's a real grace and elegance and you see the astronauts learn to operate in this completely new environment. </s><s>[INST] What are some interesting differences between the parabolic plane and when you're actually going up in orbit? Is it that with orbit you can look out and see that blue little planet of ours, [/INST] you can see the blue marble, the stunning overview effect, which is something I hope to see one day. Um What's also really different is if you're in orbit for any significant period of time, it's gonna be a lot more physiological changes to your body than if you just did an afternoon flight on the vomit, comet, everything from your bones, your muscles, your eyeballs change shape. Uh There's a lot of different things that happened for long duration spaceflight. And we still have to, as scientists, we still have to solve a lot of these interesting challenges to be able to keep humans thriving in microgravity or deep duration space missions. </s><s>[INST] Deep duration space missions. Okay let's talk about this. Um I was just gonna ask a bunch of dumb questions. So approximately how long does it take to travel to mars asking for a friend [/INST] asking for a friend about three years for a round trip and that's not that it actually </s><s>[INST] why the round trip is that? Well you're just asking [/INST] it's </s><s>[INST] Okay cool. So for just like literally flying to Mars and around it takes three years [/INST] there's some interstitial time there because you really can only go between Earth and mars at certain points in their orbits Where it's favorable to make that journey. And so part of that three years is you take the journey to Mars a few months, 6-9 months You're there for a period of time until the orbits find a favorable alignment again and then you come back another 6-9 months. </s><s>[INST] So one way travel 69 months they hang out there on vacation and come back [/INST] for vacation. </s>

<s>[INST] knowledge. So is there some sense in which it's okay to make suboptimal quote unquote moves [/INST] you have to, I mean you have to because the best moves have been analyzed to death mostly. So </s><s>[INST] that's kind of when you say something bluff, that's a kind of sacrifice. You're, you're sacrificing the optimal move, the optimal position so that you can take the opponent. I mean that's a game theoretic, you take the opponent to something they didn't prepare. Well [/INST] yeah, but you could also look at it another way that regardless, like if you turn on whatever engine you turn on, like if you try to analyze either from the starting position or the starting position of some popular opening. Like if you analyze long enough it's always going to end up in a draw. So in that sense you may not be going for like the objective, the trials that are objectively the most difficult to draw against. But you know you are trying to look at least at the less obvious path. </s><s>[INST] How much do you use engines to use Leila stock fish in your preparations? [/INST] My team does personally, I try not to use them too much on on my own because I know that when I play you can obviously cannot have help from from engines and often I feel like often having imperfect or knowledge about a position or some engine knowledge can be a lot worse than than having no knowledge. So I try to look at engines as little as possible. </s><s>[INST] So yeah, so your team uses them for research for generation of ideas but you are relying primarily on your human resources. [/INST] Yeah, for sure. You </s><s>[INST] can evaluate well, you don't leave [/INST] and I can evaluate as a human. I can know what they find unpleasant and, and, and so on and it's very often the case for me to some extent, but a lot for for others that you arrive in a position and your opponent plays a move that you didn't expect and you know, if you didn't expect it, you know that it's probably not a great move since it hasn't been expected by by the engine, but if it's not, if it's not obvious why it's not a good move, it's usually very, very hard to figure it out. And so then looking at the engines doesn't necessarily help because at that point like you're facing a human, you have to sort of think as a human, </s><s>[INST] I was chatting with the demonstrates Abbasi of deepmind a couple of days ago and he asked me to ask you about what you first felt when you saw the play of Alpha Zero, like interesting ideas, any creativity um did you feel fear that the machine is taking over? Did you, were you inspired? And and what, what was going on in your mind and heart? [/INST] Funny thing about them is, is he he doesn't play chess at all? Like, like an Ai he plays in a very, very human way? No, I was hugely inspired when I, I saw the games at first. Um and in terms of man versus machine, I mean that battle was, was kind of lost for humans even before I enter top level chess. Um so that's never been an issue for me. I never, never liked playing as computers much anyway, so, so that's completely fine. But it was amazing to see how they, quote unquote, thought about chess and in such a different way and in a way that you could mistake for creativity, </s><s>[INST] mistake for strong words. Uh is it well to you, how many sacrifices? It's willing to make that like sacrifice pieces and then wait for prolonged periods of time before doing anything with that. Is that is that weird to you, that that's part of chess? [/INST] No, it's one of the things that's hardest to replicate as a human as well, or at least from my playing playing style that usually when I, I sacrifice, I feel like I'm, you know, I don't do it unless I feel like I'm getting something like tangible, uh, in, in return and like </s><s>[INST] a few moves down the line. [/INST] If you move down the line, you can see that you can either retrieve the material or you can put your opponent's king under pressure or have some very like very concrete positional advantage that sort of compensates for it for instance in chest. So bishops and knights are fairly equivalent. Um, we both give them three points, but bishops are a little bit better and especially a bishop pair is a lot better than, than a bishop and a knight. So, or especially two nights depends on the position, but like on average they are so like sacrificing a pawn in order to get, get a bishop pair. That's one of the most common sacrifices in. </s><s>[INST] Yeah, [/INST] I mean it depends on the situation, but generally that's fine. And there are a lot of openings that are based on that, that you sacrifice a pawn for the bishop pair and then eventually it's some sort of positional equality. Um, so that's fine. But the way alpha zero would, would sacrifice a night or sometimes £2.03 ponds </s><s>[INST] and you can [/INST] see that it's looking for some sort of positional domination, but it's, It's hard to understand and it's um, it was really fascinating to see. Um, yeah, in 2019 I was sacrificing a lot of lot of ponds especially and it was, it was a great joy. Unfortunately, it's not so easy to continue to do that. People, people have found a more solid opening lines since that don't allow me to, to do that as often. I'm still trying both to get those positions and still trying to, to learn the art of, of sacrificing pieces. </s>

<s>[INST] and they just have, I mean the best of enterprise probably just has a very large dataset or deep learning can probably shine. [/INST] That's correct. That's right. And then they, I think the other pieces that they weren't again with two point or the developer summit we put together is the whole tensorflow extended piece, which is the entire pipeline. They care about stability across doing their entire thing. They want simplicity across the entire thing. I don't need to just train a model, I need to do that every day again over and over again. </s><s>[INST] I wonder to which degree you have a role in. Uh, I don't know. So I teach a course on deep learning, I have people like lawyers come up to me and say, you know, say one is machine learning going to enter legal legal realm the same thing and um, all kinds of disciplines, uh immigration insurance. Often when I see what it boils down to is these companies are often a little bit old school and the way they organized the day. So the data is just not ready yet. It's not digitized. Do you also find yourself being in the role of and an evangelist for like uh let's get organized your data folks and then you'll get the big benefit of tensorflow. Aw, do you get those? Have those conversations? [/INST] So yeah, you know, I get all kinds of questions there from, Okay, what can I, what do I need to make this work right to really need deep learning. I mean they're all these things. I already use this linear model. Why would this help? I don't have enough data or let's say, you know, or I want to use machine learning to have nuclear where to start. So it really starts to all the way to the experts who wise for very specific things. It's interesting </s><s>[INST] is there a good answer. It boils down to. Oftentimes digitizing data. So whatever you want, automated, whatever day you want to make predictions based on, you have to make sure that it's in an organized form you like with within intensive for ecosystem, there's now you're providing more and more data sets and more and more pretrained models. Are you finding yourself also the organizer of datasets? [/INST] Yes, I think the tensorflow datasets that we just released that's definitely come up where people want these datasets. Can we organize them and can we make that easier? So that's, that's definitely one important thing. The other related thing I would say is I often tell people, you know what, I don't think of the most fanciest thing that the newest model that you see make something very basic work and then you can improve it. There's just lots of things you can do with their </s><s>[INST] Yeah, I started with the basic truth. One of the big things that makes it makes sense of flow even more accessible was the appearance whenever that happened of caress the care is standard sort of outside of tensorflow. I think it was carrots on top of piano at first only and then uh paris became on top of tensorflow. Do you know when uh paris chose to also add tensorflow as a back end. Who was the, was it just the community that drove that initially? Do you know if there was discussions, [/INST] conversations? Yeah. So Francis started the campus project before he was at google and the first thing with piano. I don't remember if that was Africans flows created away before and then at some point in tensorflow started becoming popular, there were enough similarities that you decided to okay create this interface and and pretends flows back in um, I believe that might still have been before he joined google. So I, you know, we weren't really talking about that. He decided on his own and I thought that was interesting and relevant to the community. In fact, I didn't find out about him being a google until a few months after he was here. He was working on some research ideas and doing caress and his nights and weekends project and </s><s>[INST] so interesting. So he wasn't like part of the tensorflow, he didn't join, she [/INST] joined research and he was doing some amazing research, has some papers on the iron research he's done, he's a great researcher as well. And at some point we realized, oh he's he's doing this good stuff. People seem to like the FBI and he's right here. So we talked to him and he said, okay, why don't I come over to your team and work with you for a quarter and let's make that integration happen. And we talked to his manager and he said sure, my quarters fine. And that quarter has been something like two years now and so he's fully on this. </s><s>[INST] So chara's got integrated into tensorflow like in a deep way And now with 2.0 Tensorflow 2.0, sort of caress is kind of the recommended way for a beginner to interact with which makes that initial sort of transfer learning or the basic use cases even for enterprise? Uh Super simple. Right? So what what was that decision like? That seems like uh uh it's kind of a bold decision uh as well, we [/INST] did spend a lot of time thinking about that one, we had a bunch of api somebody by us, uh there was a parallel layers api that we were building and when we decided to do keras in parallel, so they were like okay, two things that we're looking at, and the first thing we was trying to do is just have them look similar, like being as integrated as possible, share all of that stuff. They were also like three other mps that others had built over time because we didn't have a standard one. Um but one of the messages that we keep kept hearing from the community, okay, which one do we use? And they kept saying like, okay, here's a model in this one and here's a model in this one which is like pick. Um so that that's sort of like, okay, we had to address that straight on with two point. Oh, the whole idea is you need to simplify, you had to pick one based on where the where you're like, okay let's see what what are, what are the people like and carries was clearly one that lots of people loved, there were lots of great things about it. Uh so pretty settled on night, </s>

<s>[INST] So it's almost like a phase transition of some [/INST] kind. It's exactly it it's a phase transition, but it's a double phase transition. It turns out it happens. I mean there's actually two things going on at once at this phase transition, which is uh which is very remarkable about. Okay, so, so uh a lot of the most important algorithms are based on random processes. And so I wanted to, you know, I want to understand random processes now. So there are data structures that sort of grow this way. Okay, so, so, so Dick Karp, one of the leading experts on random randomized algorithms, had his students working looking at this at Berkeley and we heard a rumor that the students have found something interesting happening. You know, the students are generating this, are simulated this random evolution of graphs and undertaking snapshots. Uh It was so often take a look at what the graph is. And the rumor was that every time they looked there was only one component that had loops in it. Almost always, they do a million experience. And uh only three or four times did they ever happened to see a loop at at this point? Uh no more than one component with the loop. So they're watching until the graph gets completely full. Uh So it starts out totally empty and gets more and more more and more edges all the time. Uh And and so Ok, certainly a loop comes along once, but now all the loops stay somehow joined to that one. There never there never were two guys with loops </s><s>[INST] uh [/INST] in these experiments. Okay, so anyway, this almost always. Not always, but but with very high probability that seemed to be true. So so we heard about this rumor at stanford and we said if that's true then must you know, a lot more must also be true. So there's a whole bunch, there's a whole theory out there waiting to be discovered that we have ever thought about. So let's take a look at that. And so we looked closer and we found out no, actually it's not true, but but in fact it's almost true. Namely there's a very short interval of time what is true and if you don't happen to look at during that short interval of time then you miss it. So in other words there will be a period where there two or three components have loops but they joined together pretty soon. </s><s>[INST] So [/INST] if you don't have a real fast shutter speed, you're gonna miss you're going to miss that instant. </s><s>[INST] So separate loops don't exist for long. [/INST] That's it. Yeah, I started looking at this to make it quantitative and uh the basic problem was to slow down the big banks so that I could watch it happening. I think I can explain it actually in fairly elementary terms. Even without writing a formula, like talking would do uh and and so uh that's a let's watch the evolution. And at first uh these edges are coming along and they're just making things without loops, which we call trees. Okay. So then all of a sudden the loop first appears. So at that point I have one component that has a loop. All right now, now I say that the complexity of a component is the number of edges minus the number of vertex is so if I have a loop, I have like a loop of length five has five edges and five vortices. Um or I could put a tail on that and that would be another edge. Another vertex, </s><s>[INST] 0, 1, 2 complexity kind of thing. So [/INST] if the if the complexity is zero, we have 11 loop, I call it a cycle or I call the cyclic components. So cyclic component looks like a wheel to which you attach fibers or trees, they go branching. But there's no more loops. There's only one loop and everything else feeds into that loop. Okay. And that has complexity zero. But the tree itself has complexity minus one because it has uh like like like it might have 10 versus season and nine edges to tie the time together. So nine minus 10 is minus one. So complexity -1 is a tree. It's got to be connected. That's what I mean by a compound. That's got to be connected. So, so, if I have 10 things connected, I have to have nine edges. </s><s>[INST] Can you clarify why when complexity goes? Uh can go about zero? I'm a little. Yes. [/INST] What? Right, so the complexity plus one is the number of loops. So, if complexity is zero, I have one loop. If if complexity is one, that means I have one more edge than I have heard it. So, I might have like 11 edges and 10 vortices. So it turns we call that a bicycle because it's got two loops and it's gonna have two loops. And if well, </s><s>[INST] why can't you be trees just going out for the loop? [/INST] That I would need more edges than All </s><s>[INST] right, okay, So [/INST] every time I get another loop I get another excess of edges over verte asi Ok, so in other words, uh we start out and after I have one loop, I have one component to ask a cycle in it. Now, the next step, uh according to the rumor would be at the next step, I would have a bicycle. Uh in the evolution of almost all graphs, it would go from cycle two bicycle, but in fact, there's a certain probability it goes from cycle to to, you know, 2 2 different cycles. All right. Um and I worked out the probability was something like five out of 24, there was it was substantial. Uh but still uh soon they're going to merge together almost. Okay. So but then it splits again after you have either Either two or 1 one. Uh the next step is you either have three or you have to one or you have 111. Okay. And so I worked out the probability for those transitions and I worked it out up to up to the first five transitions. And I had these had these strange numbers 5 24 and I stayed up all night and about three a.m. I had the numbers computed and I looked at him and here we're the denominator was something like 22, So the probability was something over two, </s>

<s>[INST] I'm a silly little kid trying to do a bit of good in this world. Thank you for having my back through all of it, all of my mistakes. Thank you for the love. And now, a quick few second mention of each sponsor. Check them out in the description, it's the best way to support this podcast. We got House of Macadamia's for snacks, Linode for Linux systems, Onit for supplements, Inside Tracker for bio monitoring, and ExpressVPN for security and privacy. Choose Wise and my friends. And now, on to the full ad reads. Never any ads in the middle, I hate those, I tried to make these interesting but if you [/INST] skip them, please still check out our sponsors, I enjoy their stuff, maybe you will too. </s>

<s>[INST] will also use it and do the same thing. So you have to be, you have to avoid using the things that people that are exploiters, manipulators use, because it might signal to others [/INST] that this person is trying to trick me. </s><s>[INST] Gotta be very conscious of it, yeah. What's labeling that you mentioned? The thing you like. [/INST] Well, I said earlier that old progression from you sound angry to this is probably gonna make you angry, to I know you're angry. Labeling is hanging a label on an emotion or an affect, </s><s>[INST] and then just calling it up. Is that almost always good? Could it be a source of frustration when a person is being angry and you kind of put a label on it, call out the elephant? Is it possible that that will lead to escalation of that feeling versus a resolution? [/INST] Well, the con, what would make it bad? Like, if I'm pointing out like that blatantly obvious, like if I say, look, I need you to get up and go down to the bank and make the deposit. Let's say I'm talking to somebody who works at my company. I need you to get on the phone with this person and make the appointment. And they go, sounds like you want me to talk to this person. That would be annoying. If it's just so absurdly obvious that there's no insight in your label at all. And as soon as you're demonstrating an awareness or a subtlety or an insight, either to you or to them, now we're making progress. So the only time a label could ever potentially be counterproductive is like, if you weren't actually listening </s><s>[INST] and the label indicates that you're not listening, [/INST] indicates that you're, you know, I'm teaching USC and I'm teaching labels and one of the kids in a class, he just wants to take the skills and make his deals and just hustle them. And he's just looking for a hustle. So he writes up a paper about, you know, he goes, there's some malls by Palm Springs or someplace, some platform malls. A lot of people go to buy suits. So he goes in there and he immediately starts the bargaining that my book teaches with no empathy. And he's like, throws a price at a guy and a guy's like, no. And he throws another price at a guy, guy's like, no. And then he says to the guy behind the counter, sounds like we can make a deal. Like, no, it doesn't. I just shot down everything you just said. If anything, it sounds like we're never going to make a deal. But he tried to use this label for manipulation. Now the guy didn't </s><s>[INST] get mad on the other side, but it's like clearly this dude is not listening. And at the core of everything, you have a bunch of like, you know, almost like hacks, like techniques you can use, but at the core of it is empathy. At the core of it is empathy. Yeah. That's the main thing, [/INST] you can be able to just sit there and listen. Yeah, and perceive. Yeah, and look for insights. </s><s>[INST] You know what? I like silence. Or like you're both sitting there, chilling with a drink, looking up at the stars. There's a moment, the silence makes you kind of zoom out, realize you're in this together, as opposed to playing a game or some kind of like chess game of negotiation you're in it together. I don't know. There's some intimacy to the silence. And like, if I ask a question and just let the other person sit there in silence before they answer or vice versa. They ask me a question, I sit there in silence. That's a big, it feels like [/INST] a big intimate thing. Yes. And the other two types until they've experienced that are afraid of it. And what I'm actually going to do is for whatever reason, I'm really comfortable with silence. I think because I've experienced its effectiveness and also my son Brandon, like he's a king of dynamic silence. Like he coaches people, he says, go silent, count thousands to yourself. Don't stop till you run out of numbers. That's a good line. He's also good, full of good lines. He is. That is. And so there's so much to it. But the other two types are natural wiring against it until they've experienced it. And you've got instinct, intuition has given you data once you've experienced it. But your amygdala is kicking into gear. Again, sorry, I realize it's more complicated than that until you've experienced it. So accommodators, hope-based, how do they signal fury? The silent treatment. So when you go silent, they're scared to death, you're furious because that's how they indicate it. The assertive thinks that you as the analyst went silent because you want them to talk some more. When a point of fact, you're thinking or, and I love your description, the feeling of intimacy and silence and experiencing the moment, because I'm actually going to factor that into trying to get the accommodators to love shared intimacy. They would love to experience a moment. And I can see that being very compelling than be willing to cross that chasm and experience silence and see how it works for them. </s>

<s>[INST] Anyway, click the magic spoon dot com slash lex link in the description and use code lex at check out for free shipping to let them know I sent you and also indirectly to make my dad happy. This show is also sponsored by Express VPN. Get it at express VPN dot com slash lex pod. They gave me a suggested opening line of using the internet without Express VPN is like going to the bathroom and not closing the door. This is like gpt three suggesting to me how to be more human like and I'll honestly take all the help I can get by way of life advice. Let me tell you that you need a VPN to protect you from Russians like me. In fact, this podcast is a kind of hack of your biological network where I use my monotone, low energy voice to convince you to buy a kind of expensive cereal as a way to influence the stability of the US economy. I use express VPN on both windows and Linux to protect myself if I ever do shady things on the internet, which of course I never do and never will so secure your online activity by going to express VPN dot com slash flex pod to get an extra three months free and to support this podcast. And now here's my conversation with Manila's callous [/INST] what </s><s>[INST] is beautiful about the human genome? [/INST] Don't get me started. So first of all, as an engineering feat, the human epic genome manages the most compact, the most incredible compaction you could imagine. So Every single one of your cells contains two m worth of DNA. And this is compacted in a radius which is 1000th of a millimeter. That's six orders of magnitude. To give you a sense of scale. It's as if a string as tall as the birds are khalifa, which is about a kilometer tall, was compacted into a tiny little ball the size of a millimeter. And if you put it all together, if you stretch the trillions of cells that we have, we have about 30 trillion cells in your body. If you stretch the DNA, the two m worth of DNA in every one of your trillion cells, you would basically reach all the way to jupiter 100 times. </s><s>[INST] Yeah, it's all curled up in there. It's 30 trillion cells. [/INST] 30 trillion human body, every one of them two m worth of DNA. So all of that is compacted through the epic gino. The ePI genome basically has the ability to compact this massive amount of D. N. A. From here to jupiter 10 times into one human body into just the nuclei of one human body. And the vast majority of human bodies not even need these nuclei. And that's sort of the structural part. So that's the boring part. That's the structural part. The functional part is way more interesting. So functionally what the human effigy gnome allows you to do. It's basically control the activity patterns of thousands of genes. So 20,000 genes in the human body, every one of your cells only needs a few 1000 of those but a different few 1000 of those and the way that your cells remember what their identity is is basically driven by the epa gino. So the opinion is most structural in making this dramatic compaction and it's also functional in being able to actually control the activity patterns of all yourselves. </s><s>[INST] Now can we draw a definition distinction between the genome and the ePI genome? [/INST] Again being greek, EPI means on top of so the genome is the D. N. A. And the epic Gino means anything on top of the DNA and there's three types of things on top of the DNA. The first is chemical modifications on the DNA itself. So we like to think of four bases of the D. N. A. A. C. G. T. C. Has a metal form Which is sometimes referred to as the 5th base. So metal see takes a different meaning. So in the same way that you have annotations in an orchestra score that basically say whether you should play something softly or loudly or space it out or you know, uh interpret basically the score, the human epidemiologic you to modify that primary score. So a modified C basically says play this one softly. It's basically a sign of repression in a gene regulatory region. </s><s>[INST] I love how you're talking about the function that emerges from the epa genome as a as a musical score. [/INST] It is in many ways and uh every single cell plays a different part of that score. It's like having all of human knowledge in 23 volumes like 23 giant books which are your chromosomes and every single cell has a different profession. a different role. Some cells play the piano and they're looking at chapter seven from chromosome 23 chapter four from chromosome two and so and so forth. And each of those uh pieces are all encoding in the same D. N. A. But what the epi genome allows you to do is effectively conduct the orchestra and coordinate the pieces so that every instrument plays. Only the things that it needs to play. </s><s>[INST] one thing that kind of blows my mind. Maybe you can tell me your thoughts about it is the way evolution works with natural selection is uh based on the final sort of the entirety of the orchestra musical performance. Right? And then but there's these incredibly rich structural things like each one of them doing their own little job that somehow work to get. Like the evolution selects based on the final result. And yet all the individual pieces are doing like infinitely miniscule specific things. How the heck does that work? [/INST] It's a very good insight. And you can even go beyond that and basically say evolution doesn't select at the level of an organism. It actually selected the level of whole environments called ecosystems. So let me break this down. So you basically have at the very bottom every single nucleotide being selected. But then that nucleotides function is selected at the level of, you know, each gene and every, not even each gene, each gene regulatory control element. And then those control elements are basically converging onto the function of the gene. And many genes are converging onto the function of one cell. And many cells are converging into the function of one tissue or organ and all of these organs are converging onto the level of an organism. But now that organism is not in isolation. So if you basically think about why is altruism, for example a thing, why are people being nice to each other? It was probably selected and it was probably selected because those species that were just nasty to each other didn't survive as a species. And now if you think about um symbiosis of you know uh there's plans for example that loves CO. Two and there's humans that love 02. And we're sort of you know trading um different types of gasses to each other. If you look at um ecosystems where one organism was just really nasty that organism actually died because everyone they were being nasty to was killed off and then that kind of you know, universe of life is gone. So basically what emerges is selection at so many different layers of benefit including all of these nucleotides within a body interacting for the emergent functions at the body level. Yeah. I </s>

<s>[INST] up and then there will be these great speeches at the Supreme Court by saying oh sure you can erase the mind of the robot just like you can erase the mind of a human. We both can suffer. There'll be some epic like Obama type character with a speech that we we like the robots and the humans are the same. We can both suffer. We can both hope we can both all those all those kinds of things raise families, all that kind of stuff. It's it's interesting for these just like you said emotions seems to be a Fascinatingly powerful aspect of human human interaction, human robot interaction. And if they're able to exhibit emotions at the end of the day that's probably going to have us deeply consider human rights like what we value in humans, what we value and other animals. That's why robots and ai is great. It makes us ask [/INST] really good questions. The hard questions. Yeah but you asked about you asked about the chinese room type argument. You know is it real? If it looks real? I think the chinese room argument is a really good this one. So </s><s>[INST] so so for people who don't know chinese rumors uh you can you can I don't even know how to formulate it well but basically you can mimic the behavior of an intelligence system by just following a giant algorithm code book that tells you exactly how to respond in exactly each case. But is that really intelligent? It's like a giant look up table when this person says this you answer this when this person says this you answer this. And if you understand how that works, you have this giant, nearly infinite look up table. Is that really intelligence? Because intelligence seems to be a mechanism that's much more interesting and complex than this. Look up [/INST] table. I don't think so. So the I mean, the real question comes down to do you think ah you know, you can you can mechanize intelligence in some way, even if that involves learning. And the answer is of course, yes. There's no question. There's a second question then, which is assuming you can reproduce intelligence in sort of different hardware than biological hardware. You know what our computers. Um can you, you know, match human intelligence in all the domains in which humans are intelligent? Is it possible? Right? So this is quite the hypothesis of strong ai the answer to this in my opinion is unqualified. Yes. This will this will happen at some point. There's no question that machines at some point will become more intelligent than humans in all domains where humans are intelligent. This is not for tomorrow, it's going to take a long time regardless of what, you know, Iran and others have claimed or believed. Um this is a lot a lot harder than many of many of those guys think it is. Um and many of those guys who thought it was simpler than that years, you know, five years ago now. I think it's hard because it's been five years and they realize it's going to take a lot longer. That includes a bunch of people at Deepmind for example. But um </s><s>[INST] interesting, I haven't actually touched base with the deepmind folks, but some of it Ellen or uh I mean sometimes in your role you have to kind of create deadlines that are nearer than farther away to kind of create an urgency because you know, you have to believe the impossible is possible in order to accomplish it. And there's of course a flip side to that coin. But it's, it's a weird, you can't be too cynical if you want to get something [/INST] done. Absolutely, I agree with that. But um, I mean you have to inspire people right to work on sort of ambitious things. Um, so you know, it's, it's uh it's certainly a lot harder than we believe, but there's no question in my mind that this will, this will happen. And now, you know, people are kind of worried about what does that mean for humans? They are going to be brought down from their pedestal, You know, a bunch of notches with that and uh, you know, is that going to be good or bad? I mean it's just going to give more power right? It's an amplifier for human intelligence really. </s><s>[INST] So speaking of doing cool ambitious things Fair, the Facebook Ai Research Group has recently celebrated its 8th birthday or maybe you can correct me on that. Looking back what has been the successes, the failures, the lessons learned from the eight years affair. And maybe you can also give context of where it is the newly minted meta AI fit into how does it relate to [/INST] fair? Right. So let me tell you a bit about the organization of all this. Uh Yeah, Fair was created almost exactly eight years ago. It wasn't called Fair yet. It took that name a few a few months later. Um and at the time I joined facebook, um there was a group called the AI group that had 12 engineers and a few a few scientists like, you know, 10 engineers and two scientists or something like that. I run it for 3.5 years as a director, you know, hire the first few scientists and kind of set up the culture and organized it. You know, explained to the facebook leadership, what what fundamental research was about and how it can work within uh industry and how it needs to be open and everything. And I think it's been an unqualified success in the sense that fair has simultaneously produced, you know, top level research and advance the science and the technology provided tools, open source tools like pytorch and many others. Uh but at the same time as had a direct or mostly indirect impact on facebook at the time, no matter in the sense that a lot of systems that are that meta is built around now are, are based on research projects that started at at fair. So if you were to take out, you know deep running out of facebook services now and meta more generally, I mean the company would, would literally crumble. I mean it's completely built around around the eye these days and it's really essential to the operations. So what happened after 3.5 years is that I changed rural, I became chief scientist so I'm not doing day to day management of affair anymore and more of a kind of, you know, think about strategy and things like that and I carry my, I conduct my own research, I have you know, my own kind of research group working on self supervised learning and things like this, which I didn't have time to do when I was director. Um, so now Fair is run by joelle Pineau and onto onboard together because fear is gonna split into now there's something called Fair labs which is sort of bottom up census driven research and Fair Excel which is slightly more organized for bigger projects that require a little more kind of focus and more engineering support and things like that. So Joel needs for our lab and it's the localized all over. So there's there's no question that the leadership of the company believes that this was a very worthwhile investment and what that means is that ah it's, it's there for the long run. Right? So there is uh, if you, if you want to talk in these terms, which I don't like this, there's a business model if you want. Where where fair, despite being a very fundamental research, lab brings a lot of value to the company either mostly indirectly through other groups. Um, Now, what happened 3.5 years ago when I stepped down is was also the creation of facebook Ai, which was basically a larger organization that uh, covers fair. So fare is included in it, but also has other organizations that are uh focused on applied research or advanced development of AI technology that is more, you know, focused on the </s>

<s>[INST] the ad driven model. So they rejected the address of a model. I don't know if we could. I mean, this is a difficult question. You know, if archive was supported by ads, I don't know if that's bad for archive formats. Library was supported by ads. I don't know. I don't, I'm not, it's not trivial to me. I'm unlike, I think a lot of people, uh, I'm not against advertisements, I think as when done well are really good. I think the problem with facebook and all the social networks are the way the lack of transparency around the way they use data and uh, the lack of control the users have over their data and not the fact that data is being collected and used to sell advertisements. It's a lack of transparency, lack of control. If you if you do a good job of that. I feel like it's really nice way to make stuff free. [/INST] It's like stack overflow, right? I think they've done a good job with that. Even though, as we said, like they're capturing very little of the value that they're putting out there, right? But but it makes it a sustainable company and they're providing a lot of it's a fantastic and very productive community. </s><s>[INST] Let me ask a ridiculous tangent of a question because you wrote a paper on Game of Thrones, Battle of Winterfell just as a side. Well, I'm sorry, I noticed. I'm sure you've done a lot of ridiculous stuff like this. I just noticed that particular one by ridiculous, I mean, ridiculously awesome. Can you describe the, the approach in this work which I believe is a legitimate publication? [/INST] So going back to the original, like when we were talking about the back story of papers and the importance of that. So this was actually, you know, there was uh, when the last season of the show was airing, uh, this was during a company lunch. Well, there was in the last season there's the, there's a really big battle against the forces of evil and the, you know, the forces of good and it's called the Battle of Winterfell. And um in this battle, they're like these two armies and there's a very particular thing that they have to take into account is that in the army of debt, like if someone dies in the army of the living, like that person is going to, you know, be a reborn as a soldier in the army of the debt. And so that was an important thing to take into </s><s>[INST] Account. And the initial conditions as you specify, it's about 100,000 on each side. [/INST] Exactly. I was able, I was able to like based on some images, like on previous episodes to figure out what was the size of the armies. And so what I want, what we wanted to, what we were theorizing was like how many soldiers does like a soldier on the Army of the Living has to kill in order for them to be able to destroy the army of the dead without like losing because every time uh one of the good soldiers die is gonna turn into like the other side. And so it's so we, we were theorizing that and and I wrote a couple of Differential equations and I was able to figure out that based on the size of the Army's, I think I think was the ratio had to be like 1.7. So it had to kill like 1.7 soldiers of like the army of the dead in order for them to win the battle. </s><s>[INST] Yeah, that's, that's science. It is its most powerful. And this is also somehow a pitch for uh, like a hiring pitch in a sense, like this is the kind of uh before decides to do it [/INST] exactly. Well turned out to be, you know, as for people that have watched these shows, it's like they know that every time you try to predict something that is going to happen it's going to you're going to fail miserably and that's what happened. So it was not at all important for the show but we ended up like putting that out and there was a lot of people that share that I think was some like elements of the of the show. The cast is the show that actually retweeted that and share that. At first it was </s><s>[INST] fun. I would love this kind of calculation happened like during the making of the show or the you know, I love it like in uh for example I now know Alex Garland, the director of Ex Machina and I love it. He doesn't seem to be some not many people seem to do this but I love it when directors and people who wrote the story really think through the technical details like whether it's knowing like how things, even if it's science fiction. If you were to try to do this, how would you do this? Uh like Stephen Wolfram and his son were were collaborating with the movie arrival in designing the alien language, how you communicate with aliens. Like how would you really have a math based language that could span the alien and uh being and the human being. So I love it when they have that [/INST] kind of read the martian was also big on that like the book and the movie was all about like can we actually uh is this plausible candies happen? It was all about that and that can really bring you in like the sometimes the small details. Uh I mean the guy that wrote the martian book as another book that is also filled with those like things that when you realize that okay these are grounded in science can just really bring you in. Yeah right? Like he has a book about a colony on the colony on the moon and he goes about like all the details that would be required about setting up a colony on the moon and like things that you wouldn't think about. Like the fact that um they would you know it's hard to bring like air to the moon say so they wouldn't like how do you make that bridgeable? That environment readable? You need to bring oxygen but like you you probably wouldn't be bring nitrogen. So what you do is like instead of having a an at and an atmosphere that is 100% oxygen, you like decrease the pressure so that you have the same ratio of oxygen on earth but like lowering the pressure here and so like things like water boils at a lower temperature so people would would have coffee and the coffee would be colder. Like there was a problem in this environment in the mood. So like and these are like small things in the book but I studied physics. So like when I read this, I that throws me into like tangents and I start researching that and it's like I really like to read books and watch movies when they go to that level of detail about science. I think Interstellar was one where they also consulted heavily with with the number of, I think even resulted in a couple of a couple of papers about like the black hole visualizations and um yeah, but there's and there's even more examples of interesting science around like these fantasy uh we were reading at some point, like these guys that were trying to figure out if if the tokens Middle Earth if it was round, it was like a sphere. It was like a flat based on the map based on the map and some of the references in the in the books. And so uh we actually, I think we tweeted about that. You can actually based on the distance between the cities, you can actually prove that that could be like a map of a sphere or like a spheroid and you can actually calculate the radius of that planet. </s>

<s>[INST] isolated than ever from the West Four. [/INST] Underestimating the enhanced </s><s>[INST] power of NATO, [/INST] which will now put more pressure on </s><s>[INST] Russia's borders five, [/INST] probably putting Ukraine into </s><s>[INST] NATO [/INST] six, underestimating the damage to its own economy and certainly creating more </s><s>[INST] internal resistance in Russia seven, creating a major [/INST] readjustment of power </s><s>[INST] in its oligarch. Class eight, putting cluster [/INST] and vacuum </s><s>[INST] bombs into play [/INST] nine and </s><s>[INST] underestimating the power of [/INST] social media </s><s>[INST] worldwide. And you go [/INST] on for </s><s>[INST] a while giving a [/INST] much broader </s><s>[INST] picture ah [/INST] the history and the geopolitics of all of </s><s>[INST] this. So now, [/INST] a little bit later, </s><s>[INST] two months later, [/INST] um, what </s><s>[INST] are your thoughts about the invasion of Ukraine? [/INST] Well, it's very hard to </s><s>[INST] be honest [/INST] in this regard because the the West, the West has brought down a curtain here and anyone who questions uh, the invasion of Ukraine and its consequences is is an enemy of the people. But it's it's become so difficult. We I've never seen in my lifetime ever such a wall of propaganda as I've seen in the West. And that </s><s>[INST] includes France too because [/INST] I was there recently in England. England is of course </s><s>[INST] really [/INST] vociferous. It's it's it's shocking to me how quickly europe moved in this direction and that includes Germany. I have german friends who expressed </s><s>[INST] to me their shock [/INST] over Ukraine. I have italian friends. Same thing in Italy of course, has been perhaps the most understanding and compassionate of </s><s>[INST] countries. [/INST] So it's it's quite evident that there's a united and this attests to the power of the United States and of course you have Finland and Finland, which has generally been reasonable jumping and talking about joining NATO and Sweden too. Generally there's been some more restraint in the in the way in europe, that's what surprised me the most europe, how quickly they </s><s>[INST] felt [/INST] into this NATO basket, which is very dangerous for europe, very dangerous. This goes back to my idea what I was saying earlier about sovereignty. </s><s>[INST] These [/INST] countries have </s><s>[INST] don't really give me a [/INST] sense that they have sovereignty over their own countries. </s><s>[INST] They [/INST] don't feel that the european nations, I'm obviously intuition here is working. I just don't feel that they have freedom to say what they really think and they're scared to say it When the United States invaded Iraq in 2003. I remember with great in a sense, satisfaction that at least France chirac who I had not really know much about, stood up and said the United States, we're not going to join you in this expedition basically into madness! Schroeder in Germany same thing. Of course. Putin condemned the invasion and coo Putin had been an ally of the United States since 9 11, if you remember correctly, and had called Bush and they were getting along. So even Putin said, I won't go, no, don't go into Iraq, This is this is not the solution. He he didn't oppose Afghanistan, but he opposed Iraq. So chirac </s><s>[INST] and Schroeder [/INST] stood for the old europe. I'm I remember de </s><s>[INST] Gaulle Charles [/INST] de Gaulle, he was independent of the United States. Charles. De Gaulle pulled France out of NATO because he saw the dangers of NATO, which is to say you have to fight an american </s><s>[INST] war when they [/INST] say, and they put nuclear weapons on your territory in England and France and Italy and Germany </s><s>[INST] and when they do that, [/INST] you're, you're, you're hitched to this superpower and </s><s>[INST] you have no [/INST] say in what they're gonna do if they </s><s>[INST] declare war from [/INST] there and they use your territory, you're going to be involved in a, in a major conflict. I'm talking about sovereignty. Where is that sovereignty? They don't have it. And that has influenced their mindset for years now, since 1940s since. Well, de Gaulle was the 60s. He was he actually reversed the whole flow and he was, it was, I think it was Sarkozy who put France back into NATO and uh now it's Macron, I hope because he was talking to Putin would at least have an independent viewpoint that could be helpful here. But he, so he rolled it up. He may have told Putin something else. But within days he had rolled it up and gone along with the United States position, which was enforced by the United States in a very fierce </s><s>[INST] way. [/INST] The propaganda, as I say, I don't know how much time you spend in America, but it was officious and everything was anti Russia and Russia were killing all these people were shooting down uh civilians. Although there was no proof of it. There was just these are the accidents of war. But all of a sudden there was a campaign of criminality and they were talking about bringing Putin into war. Crime </s><s>[INST] trial. Well [/INST] why didn't they talk like that when Iraq was going on and bush was killing far more people or for that matter, why were they not talking about the </s><s>[INST] killings in [/INST] Donbas and Lugansk During that 2014 to 2000 </s><s>[INST] 22 period. [/INST] That is what it's a crime. There were so many people that were killed, many of them innocent, many of them in a so </s><s>[INST] what would be the way for Vladimir Putin [/INST] two </s><s>[INST] stop the killing in [/INST] Donbas without the invasion of </s><s>[INST] Ukraine. Yeah, [/INST] that's a very good question. And I've asked that several times and I don't have the, I have not talked to him since about </s><s>[INST] two years [/INST] now? Uh it's a very good question. What's the </s><s>[INST] mistakes, [/INST] what the human </s><s>[INST] mistakes and the leadership mistakes [/INST] Made. A very good question. You see what the American press has not said, and the Western press has not said, is that on February 24, was it? That was on that day when they invaded? They were the day before. If you check the logs of the european organization that was you supervising the was in the field in Ukraine, these are neutral observers. They were seeing heavy, heavier and heavier artillery fire going into uh into Donbass uh from the Ukrainian </s>

<s>[INST] The plasticity of the brain I guess is the magic hardware that and then its flexibility in all forms is what enables the learning most in Italy. Can you talk about organ IEDs? What are they and how can you use them to help us understand the brain and the development of the brain? [/INST] This is very, very important. So the first thing I'd like to say, please keep this in the video. The first thing I like to say is that an organ Oid A brain? Organ OID is not the same as a brain. Okay. It's a fundamental distinction. It's a system a cellular system that one can develop in the culture dish starting from stem cells that will mimic some aspects of the development of the of the brain but not all of it. They are very small maximum. They become about you know 4-5 mm in Diameters. They are much simpler than than our brain of course. But yet they are the only system where we can literally watch a process of human brain development unfold and by watch. I mean study it. Remember when I told you that we can't understand everything about development. Our own brain by studying a mouse. Well we can study the actual process of development of the human brain because it all happens in neutral. So we will never have access to that process ever. And therefore this is our next best thing, like a bunch of stem cells that can be coaxed into starting a process of neural tube formation. Remember the cube that is made by the emperor leon. And from there a lot of the cell types that are present within within the brain. And you can simply watch it and study. But you can also think about diseases where development of the brain does not proceed normally write properly. Think about neurodevelopmental diseases. There are many, many different types. Think about autism spectrum disorders, there are also many different types of autism. So there you could take a stem cell, which really means either a sample of blood or a sample of skin from the patient, make a stem cell and then with that stem cell, watch the process of formation of a brain organized of that person </s><s>[INST] with that [/INST] genetics, with that genetic code in it. And you can ask, what is this genetic code doing to some aspects of development of the brain. And for the first time you may come to solutions like what cells are involved in autism. </s><s>[INST] So many questions around this. So if you take this human stem cell for that particular person without genetic code, how and you try to build an organ oid, how often will it look similar? What's the uh [/INST] yeah, </s><s>[INST] reproducibility, yes. Or how much variability is the flip side of that? Yeah. [/INST] So there is much more variability in building organize then there then there is in building brain. It's really true that the majority of us, when we're born as babies, our brains look a lot like each other. This is the magic that the embryo does where it builds a brain in the context of a body. And um and there is very little variability there, there is disease of course, but in general little variability when you build an organ oid um you know, we don't have the full code for how this is done. And so in part of the organized uh somewhat built itself because there are some structures of the brain that the cells know how to make. And another part comes from the investigator, the scientist adding to the media factors that we know in the mouse, for example, would foster a certain step of development, but it's very limited. And so as a result, the kind of product you get in the end is much more reduction. Is this much more simple than what you get in vivo, it mimics early events of development as of today. And it doesn't build very complex type of anatomy and structure does not as of today um which happens is that in viva and also the variability that you see one or going to the next tends to be higher than when you compare an embryo to the next. </s><s>[INST] So okay then the next question is how hard and maybe another flip side of that expensive is it to go from one stem cell to an organ oid, how many can you build in life? Because it sounds very complicated. It's [/INST] work definitely and it's money definitely. Um but you can really grow a very high number of this organize, You know, can go perhaps I told you the maximum they become about five in </s><s>[INST] diameter. So this [/INST] is about the size of a, of a tiny tiny, you know, raising Or perhaps the seat of an apple. Um and so you can grow um 50-100 of those inside one big bioreactors which are these flasks where the media provides nutrients for the organize. So the problem is not to grow more or less of them? Um It's really to figure out how to grow them in a way that they are more and more produce a bill. For example, organize, organize. So they can be used to study a biological process because if you have too much variability then you never know if what you see is just an exception or really the rule. </s><s>[INST] So what is an organ that look like? Are there different neurons already emerging? Is there, you know? Uh Well first, can you tell me what kind of neurons are there? Are they sort of all the same? Are they not all the same as do, how much do we understand and how much of that variance, if any can exist and organize? Yes. [/INST] So you could grow, I told you that the brain has different parts. Uh so the cerebral cortex is on the top part of the brain batteries. Another region called the striatum that is below the cortex and so on and so forth. All of these regions have different types of cells in the actual brain. Okay. And so scientists have been able to grow, organize that may mimic some aspects of development of these different regions of the brain. And so we are very interested in the cerebral cortex. </s>

<s>[INST] Well, I love thinking about that kind of stuff too. I mean, you know, consciousness itself maybe, maybe. I mean it could be like you said, beings of pure energy. Um I think I think of life as just complex systems and the kind of form of those complex systems can take seems to be much larger than the particular biological systems we see here on earth. Um I have to ask a twitter question about aliens. Ready, This is for twitter. What would you expect from twitter? Can humans have sex with aliens neal Stephenson? [/INST] Um you could pass, </s><s>[INST] I asked the language question, can they communicate, can they fall in love before before sex. That's how it works. [/INST] So which question am I answering the sex or the love? </s><s>[INST] Um I mean it depends what is more fundamental to relations across. Yeah. Across intelligent species. [/INST] Yeah. I mean you know, sex can mean a lot of things. Um So I mean uh if your production, right? You know the winds in star trek in classic star trek, you had to to really suspend your disbelief to to think that Spock was half Vulcan and half human, right? Because that's just not gonna not gonna work D. N. A. Wise. Um So um so if by sex, you mean reproductive sex then? Um uh I would say no unless unless you go to a panspermia um kind of theory which is that uh you know, humans were seated onto the planet as part of a galactic uh you know ah program of some of some sort </s><s>[INST] and then we're just returning home hanging out with our [/INST] old relatives, distant cousins. Yeah. Um But that that doesn't seem, you know, it doesn't seem seem plausible. We we know that we know that humans had sex with neanderthals with Denise Evans, Denise Evans. Um So you could think of them as aliens that came from our planet. Um So um so that's a kind of data point I guess. Um But you know if you broaden your definition of sex to mean any kind of gratifying physical interaction than? Sure. </s><s>[INST] Right? Dancing. And that's that's how we get to love? Okay. And love can take many forms? Certainly take many forms. I have to ask you, um, in terms of space, just looking at where blue origin is, looking at what SpaceX is today And maybe looking out 10, 20 years out from now. Are you impressed of what's happening? We just saw William Shatner go up to space. [/INST] Yeah, I was, I was just watching his video this morning before I came here. </s><s>[INST] Are you impressed the word things stand today? [/INST] Yeah, I mean, I mean, SpaceX in particular has done things that are just unbelievable. Um, and Uh, I don't think anyone was anticipating um, 20 years ago, let's say when this all started, just the, uh, the speed with which they'd be able to um, rack up these incredible achievements if you've kind of uh, even seen a little bit of how the sausage is made. And, and so the difficulty of, of doing any kind of space travel. Um, what they've achieved is, uh, it's just, uh, is unbelievable. </s><s>[INST] What about the, maybe a question about Elon musk, even more than Jeff Bezos. He has a very kind of ambitious vision of this project that we're on as a species of becoming a multi planetary species and becoming that quickly. Yeah, as soon as possible. Landing on mars colonizing mars, what do you think of that project? There's two questions to ask first? The question is, what, what, what do you think about the project of colonizing mars? And second, what do you think about a human being? Who is so unapologetically ambitious at achieving the impossible? What a lot of people would say is impossible. [/INST] I think that colonizing MArs is the kind of, of goal that's, uh, it's easily stated. Uh, it's, um, it's catchy, it's, it's the kind of thing that can you inspire people to get involved in a way that some other programs might not. Um, so I think it's well chosen in that way. Um, I have technical questions about there's there's a problem of per clarets on the surface of mars that that's going to be big trouble. Um, and there's, there's radiation. So, and this is known, I'm uh, but </s><s>[INST] what about business questions you think because you mentioned sort of, uh, going outside of the solar system, would, would best be done for religious reasons. Um, what about colonizing mars, can you spin it into a business proposition? [/INST] It's hard to think of a resource that's on mars that could be brought back here cheaply enough to compete with. Um, uh, with stuff we could just dig out of the ground here or grow here. Um, so I, I don't know if there is a business plan for that or if it's just strictly we're going to go there and see what happens? Uh, you know, </s><s>[INST] maybe again, we need communism to kinda to get us going to give us a reason a little bit of the competition. Well, [/INST] there's plenty of people who are sufficiently excited by the colonized mars vision that they're willing to just go all in on it. Um, even if there's not a business plan behind it. Um, so, so I think it's well chosen. It's just um, I think it's probably the only, um, the only approach to take. Um, yeah, a lot of the, when, when white people came to this continent and, and started colonizing it, you know, uh, there was not a lot of coherent planning. Like what, what plans they did have turned out to be terrible plans. Um, you know, trying to come up with plans that extend decades into the future is, uh, is a waste of time </s>

<s>[INST] Oh, the okay, so there's too much into the world of toy problems. That's okay. But all right. But is there a way to find nice representative real world impactful instances of a problem on which demonstrate that algorithm is good. So this is kind of like the machine learning world, that's kind of what they had his best tries to do is find a data set like the real world and show the performance, all the, all the conferences are all focused on beating the performance of on that real world that is there. Is there an equivalent in the complexity analysis? [/INST] Not really. Um don can oof uh started to collect examples of graphs coming from various places. So he could have a whole zoo of different graphs that he could choose from and he could study the performance of algorithms on different types of graphs and </s><s>[INST] but there is really important and compelling to be able to define a class of graphs that that the the actual act of defining a class of graphs that you're interested in? It seems to be uh a nontrivial step before talking about instances that we should care about in the real world? [/INST] Yeah, it's uh there's nothing available there that would be analogous to the training set for supervised learning, you know, where you sort of assume that the world has given you a bunch of examples to work with. We don't really have that for problems for combinatorial problems on graphs and networks, </s><s>[INST] You know, there's been a huge growth a big growth of datasets available. Do you think some aspect of theoretical computer science might be contradicting my own question while saying it, But will there be some aspect, an empirical aspect of theoretical science which will allow the fact that these data sets are huge, will start using them for analysis sort of. You know, if you want to say something about a graph algorithm, you might take a social network like facebook and looking at sub graphs of that improve something about the facebook graph and be respected and at the same time be respected in the theoretical computer science community [/INST] that hasn't been achieved yet, I'm afraid. </s><s>[INST] Is that is that is that p equals np? Is that impossible? Is is it impossible to publish a successful paper in the theoretical computer science community that shows some some performance on a real world dataset? Or is that really just those are two different worlds? [/INST] Well, they haven't really come together. I would say that there is a field of experimental algorithm IX where people sometimes they're given uh some family of examples, sometimes they just generate them at random and they report on performance. Um but there's no convincing evidence that the sample is representative of anything at all. </s><s>[INST] So let me ask in terms of breakthroughs and open problems, what are the most compelling open problems to you? And would possible breakthroughs do you see in the near term in terms of theoretical computer science? [/INST] Well, there were all kinds of relationships among complexity classes that can be studied. Um Just to mention one thing I wrote a paper with Richard Lipton in 1979, where we asked the following question um if you take a problem, a combinatorial problem in mp let's say and you choose and you pick the size of the problem. I say it's a traveling salesman problem, but Of Size 52. And you ask, could you get an efficient a small boolean circuit Tailored for that size 52, where you could feed the edges of the graph and as boolean inputs and get as an output, the question of whether or not there's a tour of a certain length. And that would in other words, briefly, what you would say in that case, is that the problem has small circuits polynomial size circuits. Now we know that if p is equal to np then in fact these problems will have small circuits. But what about the converse? Could a problem have small circuits, meaning that it's that an algorithm tailored to any particular size could work well and yet not be a polynomial time algorithm that, as you couldn't write it as a single uniform algorithm, good for all sizes. </s><s>[INST] Just to clarify small circuits for problem with particular size or even further constraint? Small circuit for particular? [/INST] For no for all the inputs of that </s><s>[INST] set off of that size. Is that a trivial problem? For a particular instance of coming up an automated way of coming up with a circuit, [/INST] I guess that's that would be that would be that would be hard. Yeah. But you know, but there's the existential question everybody talks nowadays about every existential questions uh existential challenges. Um You could ask the question um does the hamiltonian circuit problem have uh a small circuit for for every size for each size? A different small circuit. In other words, could you tailor solutions depending on the size and get polynomial size </s><s>[INST] even if P is not equal to Np. Right. [/INST] And </s><s>[INST] what a fascinating. That's true. [/INST] Yeah. What we proved is that if that were possible, then something strange would happen in complexity theory. Some high level class, which I could briefly describe something strange would happen. So, um I'll take a stab at describing what I mean. Sure, let's go there. So we have to define this hierarchy in which the first level of the hierarchy is P. In the second level is Np. And what is Np? Np involves statements of the form. There exists something such as something holds. Um So uh for example, um there exists the coloring such that a graph can be called with food only that number of colors. Or there exists a hamiltonian circuit, </s><s>[INST] there's a statement about this [/INST] graph. So, so the np. So um and mp deals with statements of that kind that there exists a solution. No, you could imagine a more complicated uh expression which which says for all X. There exists a Y. Such that some uh proposition holds involving both X and Y. So that would say for example, in game theory for all um strategies for the first player, there exists a strategy for the second player such that the first player wins, that would be that would be at the second level of the hierarchy. The third level would be there exists in a such that for all be there exists to see but something holds and you can imagine going higher and higher in the hierarchy and you would expect that the class, the complexity class classes, the correspond to those different cases would get bigger and bigger. They </s>

<s>[INST] do you think the lot of the sort of it's a political movements that are just using? [/INST] Yeah, because what happened back in those days, it is true. But to their credit, Communist Socialists were fighting for equal rights. They were fighting for the rights of black people to have equal rights. So of course, I could see why one could say, especially in those times, you've been lynched bodies burnt, body parts showcase that window stores. Meanwhile, in africa under colonization, in your own country, in your own land and you have this group. But saying, we your fight is part of what we fight. Of course you're gonna say aside with you, especially if this is all happening at a time where, you know, so 1945 these guys who would be the liberators of various African nations for meeting with Garvey with W E B du bois and that's where this meeting is very important. It's the fifth Pan African Congress meeting, it's very important, it's gonna be the last one, but it's the most important one, because that's when they formed their plans and and really the rallied around this concept of African emancipation, emancipation and African liberation, we're going to liberate our countries then later. So that's how all of this movement started to happen. And from there, Gandhi was already making some progress with India, you know, getting them out of british rule and all of that. So all of this was happening and I really like this whole thing was bubbling, bubbling, bubbling. You know, like there's like a new force going on and then we're arrived in the late fifties and um you know, Kruma with um you know, them with the british as well, they might manage to become to become um the their colonization is over there, the first one to go in 57. And then from there, it's what we call the independence, Is that when most Southern African nation of getting the independence is different dates. Mine, april 4th 1960. So all over. So this is happening when I think about it, you're talking 57 you're talking 60 we're like, we're like at this time now with the middle of the Cold War because we have to put things in context, we don't understand what's going on, because people today ask me why do you think, because even now when they understand, oh you're right, um it makes sense if you have no economic freedom you're gonna be poor. But why why why did they go for this? Why did they go for this And they don't understand? So that's what happened. So beginning of day of times. Pre colonial Africans were free marketeers. Free enterprise. It's pretty well recorded by someone like George that's where I got the chili thing from and get an economist and then slavery happened. Colonialism happened. And then the independence is um late fifties early sixties for most for most african most african countries. So there what you have is um But then what happened there? So I told you in 45 5th and African congress in the U. K. With the liberators of africa um under the under the leadership because he was the wise you know eldest man um Dubois was he was in his seventies back in the day so he's older than them you know and he's coming with all of his ideas and everything. So we're like so there we are now in the late fifties early sixties were starting to make progress with the independence is you know India has gone there before. So all of that is starting to happen and at that time remember they they already were being introduced to the concept of socialism Marxism all of that way before by some of these you know? Um black african american intellectuals over time who are very socialist Marxist by that time. So now we're becoming independent because I do I do independent like this because I I reckon that there's still neo communism going on. So now this is happening becoming free, but then you look around, what do you see that now most of his liberators of their nations become the presence of the nation's. Remember what I told you? Most of them have drunken the social socialist Marxist socialism, kool aid. So has these african nations become independent with the first independent governments? And um you know, presidents, most of them most of them are socialists, various forms of status type of government. And this is because at that point we had made a fatal mistake of going off saying um we are Marxist socialist because you guys fight for equal rights. So in this case there should be no communism or anything like that. So not only you have that going on and the people, So right now you had this battle of ideology going on because on one end represented by freedom and the economic, what do you call it, the economic system, They were using this capitalism and these are represented by the western nations facing off with Eastern bloc, practicing various forms of state, ism socialism, communism, various forms of statism and these two are fighting for influence. So, and we also have it's also not. So two things there one is we're at a time where remember the free market concept was almost dead, almost dead. So almost every intellectual at that time was social Marxist or market Socialist, I put it in, that's what you were. So you're in a world where it was a normal thing, it was just mainstream acceptance. So not only you have that force, but at the same time if these two forces are fighting one another, it turns out that the one representing capitalism and freedom, well, sorry, but isn't it you who enslaved us and colonized us and you're fighting with with the people who represent, you know, um supposedly people who are saying that who had been fighting for equal rights for us with us for the longest time, these are our friends and that's when we made a fatal mistake. Because while, yes, there were maybe good things um to agree on with Marxist Socialist of the times, I especially, you know, um equal rights for for all people and all of that, That's the only thing we should have among the only things we should have agreed upon their violent revolution tendencies. No way, when it comes to um the economic nonsense, No way we should not have thrown the baby out with the Bathwater. But that's what we did, and that's when we made a fatal mistake. So when we became free all of these nations and most of them started with Socialist or communist leaders, my country, Socialist, uh he was a Socialist And they stayed in power for 40 years, the 1st 40 years of our freedom years and all over the continent more or less, that's what you had and on top of that something else that the french don't know that people don't know is France with its colonies. Said you cannot not do um you have to um you have to keep the french french civil law. So we're talking about the Napoleonic civil code. Are you kidding me? So that's what happened. So the reason why I go back to BLM is while I have all the respect in the world and all the compassion in the world for people like Kruma, for people like near railway for people, all of those people of those times, the liberators of africa. While I have so much love compassion for them, I am also able to say because I got the benefit of 60 some years time and you know where you get to to do a debrief and see what worked, what didn't work. What happened? We have had the 60 years to look back and to reflect so yes, I can understand why they did what they did. I can even I can understand why they sided with these people who on the surface or at least some part of a fight was the same fight as them when it came to equal rights. I can excuse them but I will not excuse the Blm founders Because that mistake was terrible 60 some years ago today. No the blacks of today you cannot be serious about black lives Mattering and staying in the same sentence and we're gonna be socialist Marx Marxist socialist, it just doesn't work. So </s>

<s>[INST] right? And it's not at all abstract. I think this this this is completely central to everything I think about in terms of artificial intelligence. I don't know if you know about the M Nus dataset with handwritten digits and uh you know I don't smoke much weed or and you really but it certainly feels like it when I look at em Nous and think about this stuff, which is like, what's the difference between one and two and why are all the twos similar to each other? What kind of transformations are within the category of what makes a thing the same? And what kind of transformations are those that make it different? And Symmetries court to that, in fact are whatever the hell our brain is doing. It's really good at constructing these arbitrary and sometimes novel, which is really important when you look at like the test or they feel novel ideas of symmetry of like what, like playing with objects, were able to see things that are the same and not. And construct almost like little geometric theories of what makes things the same and not and how to make uh programs do that in a I is a total open question. And so I kinda stared and wonder how what kind of symmetries are enough to solve the m nist handwritten digit recognition problem and write that down. [/INST] Exactly. And what's so fascinating about the work in that direction, from the point of view of of a mathematician like me in a geometry er um is that the kind of groups and symmetries, the types of symmetries that we know of are not sufficient. So in other words, like we're just gonna keep on going into the weeds on this, </s><s>[INST] uh the deeper the better [/INST] uh you know, a kind of symmetry that we understand very well as rotation. Right? So here's what would be easy if if humans, if we recognize the digit as a one. If it was like literally a rotation by some number of degrees with some fixed one in some type face, like palatine o or something. That would be very easy to understand. It was very easy to write a program that could detect whether something was a rotation of a fixed Digit one. whatever we're doing when you recognize the digit one and distinguished from the digit to, it's not that it's not just incorporating uh one of the types of symmetries that we understand. Now. I would say that I would be shocked if there was some kind of classical symmetry type formulation that captured what we're doing when we tell the difference between the two and a three. To be honest, I think I think what we're doing is actually more complicated than that. I feel like it must be, </s><s>[INST] they're so simple these numbers, I mean they're really geometric objects. Like we can draw out 123. It does seem like it should be formalized, but that's why it's so strange. You think it's [/INST] formal Izabal when something stops being a two and starts being a three where you can imagine something continuously deforming from being a to two or three. </s><s>[INST] Yeah, but that's there is a moment. I have myself a written programs that literally morph twos and threes and so on and you watch and there's moments that you notice depending on the trajectory of that transformation, that morphing that it uh it is a three and a two. There's a hard line [/INST] weight. So if you ask people, if you show them this morph, if you ask a bunch of people do they all agree about where the transition question? Because I would be surprised. </s><s>[INST] I think so. Oh [/INST] my God, okay, we have an empirical </s><s>[INST] But here's the problem. Here's the problem that if I just showed that moment that I agreed on, that's not fair. No, but say I said so I want to move away from the agreement cause that's a fascinating actually question that I want to backtrack from because I [/INST] just </s><s>[INST] dogmatically said because I could be very, very wrong. But the morphing really helps that like the change because I mean partially because our perception systems see this, it's all probably tied in there somehow. The change from one to the other. Like seeing the video of it allows you to pinpoint the place where two becomes a three much better if I just showed you one picture. I think you you might you might really, really struggle, you might call 1/7. Like I I think there's something also that we don't often think about which is it's not just about the static image, it's the transformation of the image or it's not a static shape is the transformation of the shape. There's something in the movement that's seems to be not just about our perception system, but fundamental to our cognition, Like how we think about stuff. [/INST] Yeah. And it's and you know, that's part of geometry to and in fact, again, another insight of modern geometry is this idea that, you know, maybe we would naively think we're going to study, I don't know, you know, like Punk or we're gonna study the three body problem. We're gonna study sort of like three objects in space, moving around subject only to the force of each other's gravity, which sounds very simple, right? And if you don't know about this problem, you're probably like, okay, so you just like put it in your computer and see what they do. Well, I guess what that's like A problem that punk or I want a huge prize for like making the first real progress on in the 1880s and we still don't know that much about it. um 150 years later. I mean, it's a the mongoose mystery. </s>