Datasets:

alpindale

/

two-million-bluesky-posts

like 193

Also alpindale here uploads this dataset from his name, not from Pygmalion. So Pygmalion's policies doesn't apply to him.

This data has not been anonymized AT ALL. I've said it over there, I'll say it again here
I do not care how large or small the dataset is. I did not consent to my data being used in this way.

These reports are not spam, they are legitimate users of Bluesky making lawful requests to have their data deleted. They created the posts, and simply publishing them on the AT Protocol network does not give people the right to train on them unless they have explicitly agreed to that. Public network ≠ public domain. Regardless of the ethical quandary, the dataset creator scraped posts made by citizens of the EU without their consent, meaning the dataset is very likely in violation of the GDPR. I hope HuggingFace takes it down before the EU knocks on their door.

This data has not been anonymized AT ALL. I've said it over there, I'll say it again here
I do not care how large or small the dataset is. I did not consent to my data being used in this way.

This is literally just spam

It is not spam. Take any DID ledger credential from the "author" tab, and insert it into the following link at the position specified by <here>.

https://bsky.app/profile/<here>
example: https://bsky.app/profile/did:plc:5ug6fzthlj6yyvftj3alekpj

It is not spam. Take any DID ledger credential from the "author" tab, and insert it into the following link at the position specified by <here>.

https://bsky.app/profile/<here>
example: https://bsky.app/profile/did:plc:5ug6fzthlj6yyvftj3alekpj

Still harassment

It is not spam. Take any DID ledger credential from the "author" tab, and insert it into the following link at the position specified by <here>.

https://bsky.app/profile/<here>
example: https://bsky.app/profile/did:plc:5ug6fzthlj6yyvftj3alekpj

spam is spam regardless if it has a nice outfit on

Unethical data collection is unethical data collection regardless of if the network is publicly-accessable.

It is not spam. Take any DID ledger credential from the "author" tab, and insert it into the following link at the position specified by <here>.

https://bsky.app/profile/<here>
example: https://bsky.app/profile/did:plc:5ug6fzthlj6yyvftj3alekpj

spam is spam regardless if it has a nice outfit on

Unethical data collection is unethical data collection regardless of if the network is publicly-accessable.

??? Are we good

making lawful requests

There is no real legal ground to make such requests - this is just a spam with fake "legal" requests.

• bluesky tos doesn't apply to scraped data. If its uploaded on the internet publicly for everyone to read for free, then I can scrape it, legally. Sites can ask me to not do it, but I can ignore them and do it anyway, because they can't forbid me to scrap public data.

• Pygmalion's policies doesn't apply to alpindale since he's just posting this dataset from his name.

• funny eu laws doesn't really work with this. To work, they need a company that operates in eu and does something.
  Here we have an individual posting a dataset, not a company subjected to eu laws. If you think that eu laws magically apply to each post user made on the internet, then you're mistaken. If I'm not from eu I can take them and do whatever I want, because eu laws don't apply to me.

Also if huggingface will choose to remove these datasets, they will just be uploaded elsewhere, on the platform that follows US laws and disallow unlawful removal of content like this.

This is why ai in eu can die btw. Because companies will just move to other countries, scrape data all the want and then just limit access to the eu citizens.

Also you're gravely mistaken of you think that this dataset will be used to train serious llms. This data is useless for improving generalistic models.
The only things that this data can be used for:

• train llm that will shitpost like user on bluesky for fun
• train alghoritms for social networks like bluesky to provide better recommendations, etc.

making lawful requests

There is no real legal ground to make such requests.

• bluesky tos doesn't apply to scraped data. If its uploaded on the internet publicly for everyone to read for free, them I can scrape it, legally.

• Pygmalion's policies doesn't apply to alpindale since he's just posting this dataset from his name.

• funny eu laws doesn't really work with this. To work, they need a company that operates in eu and does something.
  Here we have an individual posting a dataset, not a company subjected to eu laws. If you think that eu laws magically apply to each post user made on the internet, then you're mistaken. If I'm not from eu I can take them and do whatever I want, because eu laws don't apply to me.

This is why ai in eu can die btw. Because companies will just move to other countries, scrape data all the want and then just limit access to the eu citizens.

Also you're gravely mistaken of you think that this dataset will be used to train serious llms. This data is useless for improving generalistic models.
The only things that this data can be used for:

• train llm that will shitpost like user on bluesky for fun
• train alghoritms for social networks like bluesky to provide better recommendations, etc.

Based

It is not spam. Take any DID ledger credential from the "author" tab, and insert it into the following link at the position specified by <here>.

https://bsky.app/profile/<here>
example: https://bsky.app/profile/did:plc:5ug6fzthlj6yyvftj3alekpj

Still harassment

I see you have changed your argument to harassment rather than spam. If you consider a group of users wanting their data removed from a dataset they did not consent to being a part of, ESPECIALLY in its still-identifiable state as "harassment", you must seldom receive criticism online. If you do not want to be criticized for your utterly moronic stance on data privacy and data collection, keep that to yourself. This has just been a lovely reminder of why I've stayed away from AI-obsessed people like I've seen in this thread, many of them seem to have the mentality of "if it exists on the internet, i can do whatever I want with it" which is simply not true.

CEO Clem Delangue , Machine Learning Librarian Daniel van Strien, and Principal Ethicist Giada Pistilli have said it is a mistake to upload Bluesky dataset to Hugging Face.

Clem: https://bsky.app/profile/clem.hf.co/post/3lbvlyphqd22r
Daniel: https://bsky.app/profile/danielvanstrien.bsky.social/post/3lbvih4luvk23
Giada: https://bsky.app/profile/giada.bsky.social/post/3lbwfa6udf22c

For those in the EU, Solicitor Simon McGarr specializes in Data Privacy and Privacy Law in Ireland. He wrote a letter that people can copy and paste that to check if Hugging Face is following Article 15 of the GDPR. https://bsky.app/profile/tupped.bsky.social/post/3lbw3ev7gp22h

Idk what's "fake" about this. Maybe you losers should grow up

Spam that's literally quotes from the people that run this website....

This has just been a lovely reminder of why I've stayed away from AI-obsessed people like I've seen in this thread, many of them seem to have the mentality of "if it exists on the internet, i can do whatever I want with it" which is simply not true.

Now you're just agreeing that you're hate ai folks.

which is simply not true.

It's is simply true and that's it. You can scrape internet completely legal. It's public.

Idk what's "fake" about this. Maybe you losers should grow up

You dont know what's fake because you don't want to know. Maybe you will just grow up and understand that scraping data is legal. You made public posts, now they are open to public and can be used by anyone, that's it.

Quotes from these people doesn't revert law/what I said here. Datasets like these are completely legal in US. Huggingface can choose to not host them, that's their choose.

How many times I need to repeat that eu laws and gdpr doesn't apply to other countries.

Spam when this is the only activity your account has....

This has just been a lovely reminder of why I've stayed away from AI-obsessed people like I've seen in this thread, many of them seem to have the mentality of "if it exists on the internet, i can do whatever I want with it" which is simply not true.

Now you're just agreeing that you're hate ai folks.

which is simply not true.

It's is simply true and that's it. You can scrape internet completely legal. It's public.

Idk what's "fake" about this. Maybe you losers should grow up

You don't want what's fake because you don't want to know. Maybe you will just grow up and understand that scraping data is legal. You made public posts, now they are open to public and can be used by anyone, that's it.

Someone doesn't know what GDPR is

Someone doesn't know what GDPR is

Ok, how many times I need to repeat that eu laws don't apply outside eu? 10? 100?

making lawful requests

There is no real legal ground to make such requests - this is just a spam with fake "legal" requests.

• bluesky tos doesn't apply to scraped data. If its uploaded on the internet publicly for everyone to read for free, then I can scrape it, legally. Sites can ask me to not do it, but I can ignore them and do it anyway, because they can't forbid me to scrap public data.

• Pygmalion's policies doesn't apply to alpindale since he's just posting this dataset from his name.

• funny eu laws doesn't really work with this. To work, they need a company that operates in eu and does something.
  Here we have an individual posting a dataset, not a company subjected to eu laws. If you think that eu laws magically apply to each post user made on the internet, then you're mistaken. If I'm not from eu I can take them and do whatever I want, because eu laws don't apply to me.

Also if huggingface will choose to remove these datasets, they will just be uploaded elsewhere, on the platform that follows US laws and disallow unlawful removal of content like this.

This is why ai in eu can die btw. Because companies will just move to other countries, scrape data all the want and then just limit access to the eu citizens.

Also you're gravely mistaken of you think that this dataset will be used to train serious llms. This data is useless for improving generalistic models.
The only things that this data can be used for:

• train llm that will shitpost like user on bluesky for fun
• train alghoritms for social networks like bluesky to provide better recommendations, etc.

No, sadly for you GDPR applies to any company (in that case that would be huggingface) hosting content produced by EU citizens, whether it is based in the EU or not. If you think huggingface, a conpany that is very concerned with its public image as responsible, whose founders are EU citizens themselves, and who has a track record of complying with legislation will do anything to protect what they themselves described as a misuse, well, you might be disappointed.

Someone doesn't know what GDPR is

Ok, how many times I need to repeat that eu laws don't apply outside eu? 10? 100?

How many times should I repeat that the GDPR applies because you scraped the data of EU citizens? It does not matter if you yourself are based outside of the EU, you are publishing this data without their consent and could be subject to litigation from the EU's courts.

Special as in self important don't put words in my mouth

Special guy doesn't know how laws work when European citizens are apart of the datasets

You don't know how law work.

No, sadly for you GDPR applies to any company (in that case that would be huggingface) hosting content produced by EU citizens, whether it is based in the EU or not. If you think huggingface, a conpany that is very concerned with its public image as responsible, whose founders are EU citizens themselves, and who has a track record of complying with legislation will do anything to protect what they themselves described as a misuse, well, you might be disappointed.

No, it doesn't apply. As I said, huggingface can choose to remove if they want to follow EU law, then we will just reupload on platform that follows US laws and disallow unlawful and unfair removal of content, that's it.

We are not from eu, we can take data publicly available on internet and do whatever we want.

Your law doesn't apply to other countries. The world isn't spinning around you.

All you can achieve is removal from huggingface because they can decide to comply to eu laws. And I repeat myself a millionth time, we will just reupload on platform that respects US law.

bluesky tos doesn't apply to scraped data. If its uploaded on the internet publicly for everyone to read for free, then I can scrape it, legally. Sites can ask me to not do it, but I can ignore them and do it anyway, because they can't forbid me to scrap public data.

You can scrape what you want, but Bluesky's TOS explicitly states that all users retain ownership over content that they publish (2D). Bluesky retains a limited license, which is extended to others under specific use cases, per Section 2D. If you want to scrape the firehose go right ahead, but you cannot legally redistribute, or create derivative works from my user content without my consent--save for exceptions under 2D. Bluesky would be the final arbiter of what does and doesn't count as an exception, but if you have a working brain, this clearly isn't one.

Even if that's not all the case (which it is lol), it's still in clear violation of CCPA for some 39 million Californians. Did minors between 13 and 16 (Bluesky is 13+) give explicit consent? CCPA requires that their consent for data collection be opt-in, as opposed to opt-out, which it still requires for all Californians above 16. On that, there is no clear way to request for data be deleted, which is also a right granted by CCPA. The CCPA also expects data of Californians to be anonymized, which this dataset is not. Honestly this dataset violates almost every right outlined in the CCPA.

There is no real legal ground to make such requests - this is just a spam with fake "legal" requests.

In absence of an actual system for requesting takedowns which is required by law as per above, I think it makes a lot of sense for them to be submitted here!

Furthermore, while not a legal issue, HuggingFace clearly doesn't love what's happening here either. On the dataset from yesterday:

• Daniel apologized for violating consent and principles of transparency: https://bsky.app/profile/danielvanstrien.bsky.social/post/3lbvih4luvk23
• Giada called the dataset "inappropriate and unethical": https://bsky.app/profile/giada.bsky.social/post/3lbwfa6udf22c
• Giada also said re:Daniel's data set "This won't be happening again in the future": https://bsky.app/profile/giada.bsky.social/post/3lbwfa6un5k2c
• Clem supported its removal: https://bsky.app/profile/clem.hf.co/post/3lbvlyphqd22r

Anyways I have a feeling all of this will fall on deaf ears, so I'll wrap up by saying if you're in CA and your user content is being redistributed here, you should know and exercise your rights by filing a formal complaint at https://privacy.ca.gov/submit-a-complaint/ccpa-complaints/ and if you're unsure whether your user content has been stolen, you can check by using https://clearsky.app to quickly grab your account's DID and searching this dataset for it. Hope this helps <333

The law applies to people in the dataset whose data you are scrapping ...

@moonlitsara

Bluesky would be the final arbiter of what does and doesn't count as an exception, but if you have a working brain, this clearly isn't one.

No, it wont. You need a working brain to understand that I can scrape anything publicly available on internet completely lawfully in the us and toa have absolutely nothing to do with it. That's why it doesn't apply.

Also, I repeat myself 100th time since you still didn't say anything against it. Huggingface operates in eu by eu laws => they remove dataset => we reupload on us website that works by us laws and doesn't operate on eu. That's it and eu laws doesn't apply to us anymore in absolutely any way.

Also fuck California, ai companies are moving from there because they push idiotic laws about compute and other stuff. Us != California.

Special guy doesn't know how laws work when European citizens are apart of the datasets

You don't know how law work.

No, sadly for you GDPR applies to any company (in that case that would be huggingface) hosting content produced by EU citizens, whether it is based in the EU or not. If you think huggingface, a conpany that is very concerned with its public image as responsible, whose founders are EU citizens themselves, and who has a track record of complying with legislation will do anything to protect what they themselves described as a misuse, well, you might be disappointed.

No, it doesn't apply. As I said, huggingface can choose to remove if they want to follow EU law, then we will just reupload on platform that follows US laws and disallow unlawful and unfair removal of content, that's it.

We are not from eu, we can take data publicly available on internet and do whatever we want.

Your law doesn't apply to other countries. The world isn't spinning around you.

All you can achieve is removal from huggingface because they can decide to comply to eu laws. And I repeat myself a millionth time, we will just reupload on platform that respects US law.

it's honestly cute that you think that but:

• as stated by others, GDPR is only one regulation making this illegal, the state of California also has similar protection, so good luck with USA hosting.
• American companies (or companies from wherever) who don't comply to GDPR are liable to enormous fines in the EU. Of course they could chose to ignore it and as result be banned from operating in the EU. You can delude yourself in thinking american companies will take that risk for the sake of your sad existence, but, again, yiu might be disappointed.

Here's documentation, I hope it helps: https://gdpr-info.eu/issues/fines-penalties/. Maybe find a better hobby though, you're clearly not very good at this one.

Nothing that was done here is illegal in US, 100%, again. It probably (not 100%) can be illegal in eu, but we will just reupload outside eu (outside huggingface of it will comply with eu laws) and that's it. EU law doesn't apply to us. By the US law, we can scrap anything we want.

American companies (or companies from wherever) who don't comply to GDPR are liable to enormous fines in the EU. Of course they could chose to ignore it and as result be banned from operating in the EU. You can delude yourself in thinking american companies will take that risk for the sake of your sad existence, but, again, yiu might be disappointed.

@lgrobol Can you please read what I repeat 100 times and think like for a second?

We will reupload to company that doesn't operate in eu. That's it. Eu can't fine companies that doesn't operate inside it. California != us.

It's ridiculous that people actually think that eu laws can somehow reach other countries companies if they don't operate in eu.

No, it wont. You need a working brain to understand that I can scrape anything publicly available on internet completely lawfully in the us and toa have absolutely nothing to do with it. That's why it doesn't apply.

Yes you can scrape anything publicly available on the internet. Scraping is different from using it. The issue is in usage of the scraped data--republishing it without consent, creating derivative works, etc. I can read any book in a public library but I can't copy-paste it onto Amazon for other people to either take for free or to purchase. These are like, two very very distinct actions. You get that, right?

Also, I repeat myself 100th time since you still didn't say anything against it. Huggingface operates in eu by eu laws => they remove dataset => we reupload on us website that works by us laws and doesn't operate on eu. That's it and eu laws doesn't apply to us anymore in absolutely any way.

California laws protects California citizens and the dataset is a republication of content created in part by California citizens unless it somehow managed to filter them out (spoiler: it didn't!). Their data falls under protection regardless for the same reasons why US-based companies have to respect the GDPR. This is so silly :)

making lawful requests

There is no real legal ground to make such requests - this is just a spam with fake "legal" requests.

• bluesky tos doesn't apply to scraped data. If its uploaded on the internet publicly for everyone to read for free, then I can scrape it, legally. Sites can ask me to not do it, but I can ignore them and do it anyway, because they can't forbid me to scrap public data.

• Pygmalion's policies doesn't apply to alpindale since he's just posting this dataset from his name.

• funny eu laws doesn't really work with this. To work, they need a company that operates in eu and does something.
  Here we have an individual posting a dataset, not a company subjected to eu laws. If you think that eu laws magically apply to each post user made on the internet, then you're mistaken. If I'm not from eu I can take them and do whatever I want, because eu laws don't apply to me.

Also if huggingface will choose to remove these datasets, they will just be uploaded elsewhere, on the platform that follows US laws and disallow unlawful removal of content like this.

This is why ai in eu can die btw. Because companies will just move to other countries, scrape data all the want and then just limit access to the eu citizens.

Also you're gravely mistaken of you think that this dataset will be used to train serious llms. This data is useless for improving generalistic models.
The only things that this data can be used for:

• train llm that will shitpost like user on bluesky for fun
• train alghoritms for social networks like bluesky to provide better recommendations, etc.

No, sadly for you GDPR applies to any company (in that case that would be huggingface) hosting content produced by EU citizens, whether it is based in the EU or not. If you think huggingface, a conpany that is very concerned with its public image as responsible, whose founders are EU citizens themselves, and who has a track record of complying with legislation will do anything to protect what they themselves described as a misuse, well, you might be disappointed.

Not really, no.

California laws protects California citizens and the dataset is a republication of content created in part by California citizens unless it somehow managed to filter them out (spoiler: it didn't!). Their data falls under protection regardless for the same reasons why US-based companies have to respect the GDPR. This is so silly :)

Well then we just go, scrap it again and publish outside of eu and California, silly :). It's that simple. California is killing itself in the ai race.

Well then we just go, scrap it again and publish outside of eu and California, silly :). It's that simple. California is killing itself in the ai race.

oh my god oh my god PLEASEEEE think about this for a second. the law applies regardless of where it's published if it contains data of the citizens of those places. go make a startup in texas and see what happens when you unlawfully use the data of CA and EU citizens. oh my god

@SpacehogSutra https://en.wikipedia.org/wiki/General_Data_Protection_Regulation

Scraping is different from using it. The issue is in usage of the scraped data--republishing it without consent, creating derivative works, etc. I can read any book in a public library but I can't copy-paste it onto Amazon for other people to either take for free or to purchase. These are like, two very very distinct actions. You get that, right?

Yes, you cant scrap books from libraries, that's unlawful, obviously.

If you upload stolen copyrighted content to Amazon or some other website, it will be taken down, obviously.

But bluesky shitpost isn't copyrighted by anyone. It's public data.
It's like if someone uploads book for free without copyright, than anyone can use it however he want.

If bluesky was copyrighted, then it will be unlawful. but it's obviously not.

American companies (or companies from wherever) who don't comply to GDPR are liable to enormous fines in the EU. Of course they could chose to ignore it and as result be banned from operating in the EU. You can delude yourself in thinking american companies will take that risk for the sake of your sad existence, but, again, yiu might be disappointed.

@lgrobol Can you please read what I repeat 100 times and think like for a second?

We will reupload to company that doesn't operate in eu. That's it. Eu can't fine companies that doesn't operate inside it. California != us.

lmao sure, good luck with that. Somehow I won't hold my breath for it, though. And remember to get some actual legal counsel ♥️.

@moonlitsara

oh my god oh my god PLEASEEEE think about this for a second. the law applies regardless of where it's published if it contains data of the citizens of those places. go make a startup in texas and see what happens when you unlawfully use the data of CA and EU citizens. oh my god

It's so ridiculous that people think like that. Are you crazy? Eu can't do shit to companies in us if they don't operate in eu. They can't take them to court because they aren't subjected to eu laws.

the law applies regardless of where it's published if it contains data of the citizens of those places

Okay, so if china will make a law that anyone that uses data of its citizens will be charged with 9191181811819191918$ fine, then do you really think that us companies will comply if they don't work in china? It's so idiotic to think so. If company doesn't operate in eu, but operates based on laws of other country, then eu can't do absolutely anything about it. The only thing that eu can do is to declare a war to that country maybe, if it doesn't like it's laws.

lmao sure, good luck with that. Somehow I won't hold my breath for it, though. And remember to get some actual legal counsel ♥️.

Thanks for the kind words, we will absolutely do that! Looks like you run out of arguments tho, while I was saying same thing over and over again and people are so slow to understand for some reason that eu law can go fuck itself.

Though I can agree that my initial statement that these legal requests are "fake", isn't very true because huggingface operates in eu and they probably will remove this data.

But it will just be re-uploaded on the website that isn't subjected to eu laws.

(If it's really illegal in eu, because data collected there is public)

Though I can agree that my initial statement that these legal requests are "fake", isn't very true because huggingface operates in eu and they probably will remove this data.

But it will just be re-uploaded on the website that isn't subjected to eu laws.

(If it's really illegal in eu, because data collected there is public)

I hope you downloaded the model in case they bully alpindale in removing it.

Permission was given by the post authors in Section 2C of the Bluesky Terms of Service.

Permission was given by the post authors in Section 2C of the Bluesky Terms of Service.

True

Though I can agree that my initial statement that these legal requests are "fake", isn't very true because huggingface operates in eu and they probably will remove this data.

But it will just be re-uploaded on the website that isn't subjected to eu laws.

(If it's really illegal in eu, because data collected there is public)

I hope you downloaded the model in case they bully alpindale in removing it.

already cloned, why not

The irony of the anti-AI backlash is that by fighting open AI, people are actually empowering the very companies they claim to oppose. When individuals protest AI training on public data, they seem to forget that tech giants have been cataloging and owning internet content for years. Corporate AI will train regardless - the only difference is whether the process is transparent (open AI) or opaque (closed corporate systems).

The real choice isn't between AI that uses public data or not-that ship has long sailed. The choice is between collaborative, potentially democratized AI development and closed, profit-driven AI controlled exclusively by Big Tech. Attacking open AI doesn't protect your data; it just concentrates the power of AI development in fewer hands.

Will we look back on today’s anguish over open content being mined to train neural networks and marvel at how blind we were to the importance of huge, open datasets for democratic, transparent AI? Will we say that the mining of open content is vital if we want models that are less biased than those trained on smaller proprietary datasets? Will we think that open datasets are also vital if we care about barriers to entry in the new world of the data-rich and data-poor? (Remember, that may be the single most important generative inequality in the market of the next fifty years, the one from which an entirely skewed new political economy flows. And you want to make it harder to have open datasets?)

https://openfuture.eu/paradox-of-open-responses/misunderestimating-openness/

At least please be honest. As I said elsewhere, I don't care either way. But this dataset is non compliant. It could become so with a script worth 20 lines of coding, but i guess it's more important to stick it to those "anti" than being remotely fair. That's definitely one way to shoot oneself in the foot.