The ruling is fine. The judge is not Alsop but he’s not technically incompetent either, which is good.
The torrent comments in general are nothing to get het up about; in summary
1) Meta wanted to download but not upload libgen and Anna’s after they couldn’t find anyone with rights to license that would talk to them.
2) they didn't want to distribute; just download. An engineer put in evidence that they restricted seeding successfully.
3) late in the case Silverman et al claimed while they hadnt been seeding they had been leeching and that counts as distribution (?!)
Judge commented as follows
1. just downloading is probably fine because it could be for purposes of fair use, and fair use concerns generally trump even good faith and fair dealing
2. Nobody could get llama to spit out more than a 60 token quote from a plaintiff book; thus llama is not made for infringement
3. We will need more briefing on this leeching thing which it is alleged is a form of distribution.
The judge lays out what he thinks a workable claim to get to the supreme court would be, which is that these llms defeat the purpose of our copyright laws by reducing the amount of human creativity and expression available to those who want to create economic value through creativity. Eg where will the jobs for biographers go?
I will say that debate is an active topic worldwide right now and a good question, with answers ranging from: “this maximizes human creativity bro” to “laser printers disrupted lead type foundries, that was great” to “nobody will ever write again and we are murdering our creative class and burning down their craftsman mid century modern homes.”
It seems to me this will get taken up next session with SCOTUS but also that it’s a little early; we just don’t know where this is going exactly. Either way, I expect our current judge will learn that leeching is precisely NOT seeding once the defense legal team has time to brief him.
> 1. just downloading is probably fine because it could be for purposes of fair use, and fair use concerns generally trump even good faith and fair dealing
This smells a bit strange to me, it's a "for-profit" company.. Fair use is a bit of pipe-dream here. Also there is no conditions on the source of the content ? If the source was obtained from illegal sources IE illegal distribution of copyrighted materials does that not play a part ?
Also will this set a precedent that if I download HBO's collection but don't seed or use for any commercial reasons it will be considered Fair Use ?
This whole thing just reeks of "rules for thee but not for me".
> This smells a bit strange to me, it's a "for-profit" company.. Fair use is a bit of pipe-dream here
Why do you think that? For-profit companies use fair use all the time. Its not unusual.
Yes, a usage being non-commercial can be a factor in favour of fair use, but its just one factor. Its definitely not a neccesary condition nor is it a sufficient condition.
> If the source was obtained from illegal sources IE illegal distribution of copyrighted materials does that not play a part ?
Why would it? That isn't really how copyright works. Its about the right to "copy" (or not to), not about distribution methods.
> Also will this set a precedent that if I download HBO's collection but don't seed or use for any commercial reasons it will be considered Fair Use ?
No. That's not the reason this is potentially fair use.
[Although as an aside it uses to be in Canada that only uploading was illegal].
I'm not sure how llms count as fair use.
It's just that we can't show HOW they've been encoded in the model, means it's fair use? Or that statistical representations are fair use? Or is it the generation aspect? I can't sell you a Harry potter book, but I can sell you some service that let's you generate it yourself?
I feel like this has really blown a hole in copyright.
> It's just that we can't show HOW they've been encoded in the model, means it's fair use?
Describing training as “encoding them in the model” doesn’t seem like an accurate description of what is happening. We know for certain that a typical copyrighted work that is trained on is not contained within the model. It’s simply not possible to represent the entirety of the training set within a model of that size in any meaningful way. There are also papers showing that memorisation plateaus at a reasonably low rate according to the size of the model. Training on more works doesn’t result in more memorisation, it results in more generalisation. So arguments based on the idea that those works are being copied into the model don’t seem to be founded in fact.
> I can't sell you a Harry potter book, but I can sell you some service that let's you generate it yourself?
That’s the reason why cases like this are doomed to fail: No model can output any of the Harry Potter books. Memorisation doesn’t happen at that scale. At best, they can output snippets. That’s clearly below the proportionality threshold for copyright to matter.
> That’s clearly below the proportionality threshold for copyright to matter.
This type of reasoning keeps coming up with seemingly zero consideration for why copyright actually exists. The goal of copyright, under US law, is "To promote the progress of science and useful arts".
The goal of companies creating these LLMs is to supersede the use of source material they draw from, like books. You use an LLM because it has all the answers without having to spend the money compensating the original authors, or put in the work digesting it yourself, that's their entire value proposition.
Their end game is to create a product so good that nobody has a reason to ever buy a book again. A few hours after you publish your book, the LLM will gobble it up and distribute the insights contain within to all of their users for free, "it's fair use", they say. There won't be any economic incentive to write books at that point, and so "the progress of science and useful arts" will crawl to a halt. Copyright defeated.
If LLM companies are allowed to produce market substitutes of original works then the goal of copyright is being defeated on a technicality and this ought to be a discussion about whether copyright should be abolished completely, not a discussion about whether big tech should be allowed to get away with it.
Copyright was build to protect the artist from unauthorized copy by a human not by a machine (a machine wildly beyond their imagination at the time I mean), so the input and output limitations of humans were absolutely taken into account when writing such laws, if LLMs were treated in similar fashion authors would have had a say in wether their works can be used as inputs in such models or if they forbid it.
Same. If I invented a novel new way of encoding video and used it to pack a bunch of movies into a single file, I would fully expect to be sued if I tried distributing that file, and equally so if I let people use a web site that let them extract individual videos from that file. Why should text be treated differently?
You are allowed to quote from copyrighted works without needing permission. Trying to assert copyright because of a quote of, say, a mere 60 words in length would get you thrown out of any judge’s court.
It was shown, in this case, that the llms wouldn’t generate accurate quotes more than 60 words in length.
This is not comparable to encoding a full video file.
I think the better analogy is if you had someone with a superhuman, but not perfect memory read a bunch of stuff, then you were allowed to talk to the person about the things they’d read, does that violate copyright? I’d say clearly no.
Then what if their memory is so good, they repeat entire sections verbatim when asked. Does that violate it? I’d say it’s grey.
But that’s a very specific case - reproducing large chunks of owned work is something that can be quite easily detected and prevented and I’m almost certain the frontier labs are already going this.
So I think it’s just very not clear - the reality is this is a novel situation, the job of the courts is now to basically decide what’s allowed and what’s not. But the rational shouldn’t be ‘this can’t be fair use it’s just compression’. Because it’s clearly something fundamentally different and existing laws just aren’t applicable imo
People can process only fraction of that while using much more time to do that. And I'm using 'process' here to meet you in your nihilist argument that these algorithms are same as humans. Which is pretty strange because people barely acknowledge similarities with other mammals but suddenly software is equal.
I'm inclined to agree here. LLMs do not use just a paragraph here and there in accordance with fair use, but rather uses the entire body of work to train itself.
The judge is claiming that because the use is of the books are “so transformative,” the usage of these books to train an llm is fair use.
I’m not familiar with the facts of the case and IANAL, and its late, but how did the plaintiffs determine their books were being used for training of the llm? Was the model spitting out language that was similar or verbatim to their works?
>The judge is claiming that because the use is of the books are “so transformative,” the usage of these books to train an llm is fair use.
"you're doing something so critical to our (country's) success, that we're ok to waive copyright. I get that, if the US doesn't do it, then China will(is).
Interesting judgement, and it's implications, if you are correct haha.
The word transformative was put there in a time of manual transformative processes, like when you paint something similar to what you saw in a painting by another artist, with all the implied limitations that entails, like the time it took from you to watch that painting, and the time it takes you to create that new painting, nothing to do at all with the way LLMs operate, an honest assessment would have found that the word was meant for a wildly different use case and therefore it required a bigger and more nuanced discussion.
> The word transformative was put there in a time of manual transformative processes, like when you paint something similar to what you saw in a painting by another artist
Do you have any citation that that is how the word "transformation" was understood historically? Because what your suggesting seems to be the opposite of what i've read.
My understanding is even back in the 1800s (e.g. https://en.wikipedia.org/wiki/Folsom_v._Marsh ) your example would not be considered transformative, if your intention was to make a similar painting to serve a similar purpose.
That’s one hell of a headline for a story about Meta winning summary judgement for most of the claims against them. You’d be forgiven for thinking Meta lost this case, going by the headline.
It's not a story about Meta winning (that previous story is linked in the first sentence), but a story expanding on the part where they didn't. The headline is totally appropriate for that piece.
It’s crazy, yet so predictable, that while the system tries to bankrupt individuals for torrenting a single book or movie in this case the excuse “it was just to train an LLM” will fly. Imagine a private individual would argue that in court.
The system wants to destroy creativity and humans. The system wants artists and writers to depend on the patronage of the ultra rich who steal their output.
We are back to feudal society, except that monarchs or their advisers at least had taste as opposed to the Nouveau riche.
I mean it seems clear that Meta did not pirate the content to watch/read it. However, I guess according to the ruling you could pirate anything you want (but no seeding), produce a shitty haiku based on what you pirated and then claim fair use.
So the argument is that by torrenting ebooks, meta provided bandwidth to the torrent network, and thus provided (financial??!) benefit to pirate sites?
I got to be honest, that sounds extremely weak to me. The benefit to the pirate site of joining the torrent swam seems like it would be extremely slight.
The pirate sites? There is no pirate site hosting files here, BitTorrent is peer to peer.
If it’s fine for a large corp. like Meta to pirate books then it’s fine for everyone else. If it’s a crime for ordinary consumers then it’s a crime for Meta too.
Especially as Meta aren’t doing this for charity. They train LLMs for their own gain.
> The pirate sites? There is no pirate site hosting files here, BitTorrent is peer to peer.
Or "shadow library" or whatever you want to call it. The argument according to the article is that the entity that created the torrent, which also as far as i understand also operates a traditional website, benefits from meta's actions.
Just read (most) of the ruling.
The ruling is fine. The judge is not Alsop but he’s not technically incompetent either, which is good.
The torrent comments in general are nothing to get het up about; in summary
1) Meta wanted to download but not upload libgen and Anna’s after they couldn’t find anyone with rights to license that would talk to them.
2) they didn't want to distribute; just download. An engineer put in evidence that they restricted seeding successfully.
3) late in the case Silverman et al claimed while they hadnt been seeding they had been leeching and that counts as distribution (?!)
Judge commented as follows
1. just downloading is probably fine because it could be for purposes of fair use, and fair use concerns generally trump even good faith and fair dealing
2. Nobody could get llama to spit out more than a 60 token quote from a plaintiff book; thus llama is not made for infringement
3. We will need more briefing on this leeching thing which it is alleged is a form of distribution.
The judge lays out what he thinks a workable claim to get to the supreme court would be, which is that these llms defeat the purpose of our copyright laws by reducing the amount of human creativity and expression available to those who want to create economic value through creativity. Eg where will the jobs for biographers go?
I will say that debate is an active topic worldwide right now and a good question, with answers ranging from: “this maximizes human creativity bro” to “laser printers disrupted lead type foundries, that was great” to “nobody will ever write again and we are murdering our creative class and burning down their craftsman mid century modern homes.”
It seems to me this will get taken up next session with SCOTUS but also that it’s a little early; we just don’t know where this is going exactly. Either way, I expect our current judge will learn that leeching is precisely NOT seeding once the defense legal team has time to brief him.
> 1. just downloading is probably fine because it could be for purposes of fair use, and fair use concerns generally trump even good faith and fair dealing
This smells a bit strange to me, it's a "for-profit" company.. Fair use is a bit of pipe-dream here. Also there is no conditions on the source of the content ? If the source was obtained from illegal sources IE illegal distribution of copyrighted materials does that not play a part ?
Also will this set a precedent that if I download HBO's collection but don't seed or use for any commercial reasons it will be considered Fair Use ?
This whole thing just reeks of "rules for thee but not for me".
> This smells a bit strange to me, it's a "for-profit" company.. Fair use is a bit of pipe-dream here
Why do you think that? For-profit companies use fair use all the time. Its not unusual.
Yes, a usage being non-commercial can be a factor in favour of fair use, but its just one factor. Its definitely not a neccesary condition nor is it a sufficient condition.
> If the source was obtained from illegal sources IE illegal distribution of copyrighted materials does that not play a part ?
Why would it? That isn't really how copyright works. Its about the right to "copy" (or not to), not about distribution methods.
> Also will this set a precedent that if I download HBO's collection but don't seed or use for any commercial reasons it will be considered Fair Use ?
No. That's not the reason this is potentially fair use.
[Although as an aside it uses to be in Canada that only uploading was illegal].
2 replies →
> This smells a bit strange to me, it's a "for-profit" company.. Fair use is a bit of pipe-dream here
Fair use can be for profit.
> if I download HBO's collection but don't seed or use for any commercial reasons it will be considered Fair Use
No, seeding is automatically not fair use. Leeching does not automatically mean its not fair use, just that it might be.
If you pirate a movie, watch it, and talk about it but don't distribute it then yes that's fair use. People literally do this all the time.
2 replies →
I'm not sure how llms count as fair use. It's just that we can't show HOW they've been encoded in the model, means it's fair use? Or that statistical representations are fair use? Or is it the generation aspect? I can't sell you a Harry potter book, but I can sell you some service that let's you generate it yourself?
I feel like this has really blown a hole in copyright.
> It's just that we can't show HOW they've been encoded in the model, means it's fair use?
Describing training as “encoding them in the model” doesn’t seem like an accurate description of what is happening. We know for certain that a typical copyrighted work that is trained on is not contained within the model. It’s simply not possible to represent the entirety of the training set within a model of that size in any meaningful way. There are also papers showing that memorisation plateaus at a reasonably low rate according to the size of the model. Training on more works doesn’t result in more memorisation, it results in more generalisation. So arguments based on the idea that those works are being copied into the model don’t seem to be founded in fact.
> I can't sell you a Harry potter book, but I can sell you some service that let's you generate it yourself?
That’s the reason why cases like this are doomed to fail: No model can output any of the Harry Potter books. Memorisation doesn’t happen at that scale. At best, they can output snippets. That’s clearly below the proportionality threshold for copyright to matter.
> That’s clearly below the proportionality threshold for copyright to matter.
This type of reasoning keeps coming up with seemingly zero consideration for why copyright actually exists. The goal of copyright, under US law, is "To promote the progress of science and useful arts".
The goal of companies creating these LLMs is to supersede the use of source material they draw from, like books. You use an LLM because it has all the answers without having to spend the money compensating the original authors, or put in the work digesting it yourself, that's their entire value proposition.
Their end game is to create a product so good that nobody has a reason to ever buy a book again. A few hours after you publish your book, the LLM will gobble it up and distribute the insights contain within to all of their users for free, "it's fair use", they say. There won't be any economic incentive to write books at that point, and so "the progress of science and useful arts" will crawl to a halt. Copyright defeated.
If LLM companies are allowed to produce market substitutes of original works then the goal of copyright is being defeated on a technicality and this ought to be a discussion about whether copyright should be abolished completely, not a discussion about whether big tech should be allowed to get away with it.
2 replies →
Copyright was build to protect the artist from unauthorized copy by a human not by a machine (a machine wildly beyond their imagination at the time I mean), so the input and output limitations of humans were absolutely taken into account when writing such laws, if LLMs were treated in similar fashion authors would have had a say in wether their works can be used as inputs in such models or if they forbid it.
10 replies →
Same. If I invented a novel new way of encoding video and used it to pack a bunch of movies into a single file, I would fully expect to be sued if I tried distributing that file, and equally so if I let people use a web site that let them extract individual videos from that file. Why should text be treated differently?
You are allowed to quote from copyrighted works without needing permission. Trying to assert copyright because of a quote of, say, a mere 60 words in length would get you thrown out of any judge’s court.
It was shown, in this case, that the llms wouldn’t generate accurate quotes more than 60 words in length.
This is not comparable to encoding a full video file.
I think the better analogy is if you had someone with a superhuman, but not perfect memory read a bunch of stuff, then you were allowed to talk to the person about the things they’d read, does that violate copyright? I’d say clearly no.
Then what if their memory is so good, they repeat entire sections verbatim when asked. Does that violate it? I’d say it’s grey.
But that’s a very specific case - reproducing large chunks of owned work is something that can be quite easily detected and prevented and I’m almost certain the frontier labs are already going this.
So I think it’s just very not clear - the reality is this is a novel situation, the job of the courts is now to basically decide what’s allowed and what’s not. But the rational shouldn’t be ‘this can’t be fair use it’s just compression’. Because it’s clearly something fundamentally different and existing laws just aren’t applicable imo
One should also keep in mind the countless people who got much of their education from pirated books.
That has nothing to do with whether LLMs are fair use.
This seems to be a bad faith argument, although it would be amusing to see Facebook use it.
"Your honor, it's fair use because students have downloaded educational books for years."
People can process only fraction of that while using much more time to do that. And I'm using 'process' here to meet you in your nihilist argument that these algorithms are same as humans. Which is pretty strange because people barely acknowledge similarities with other mammals but suddenly software is equal.
I'm inclined to agree here. LLMs do not use just a paragraph here and there in accordance with fair use, but rather uses the entire body of work to train itself.
Or am I misunderstanding something about LLMs?
The judge is claiming that because the use is of the books are “so transformative,” the usage of these books to train an llm is fair use.
I’m not familiar with the facts of the case and IANAL, and its late, but how did the plaintiffs determine their books were being used for training of the llm? Was the model spitting out language that was similar or verbatim to their works?
> The judge is claiming that because the use is of the books are “so transformative,” the usage of these books to train an llm is fair use.
Maybe I'm mistaken but shouldn't the source come from a legal source ? This is not public domain material.
Again if I download the entire works of HBO tv shows, then make a "transformative" version on my iphone, how can that be considered fair use?
3 replies →
> but how did the plaintiffs determine their books were being used for training of the llm?
I think facebook admited this. I don't think the fact of this is under dispute.
>The judge is claiming that because the use is of the books are “so transformative,” the usage of these books to train an llm is fair use.
"you're doing something so critical to our (country's) success, that we're ok to waive copyright. I get that, if the US doesn't do it, then China will(is).
Interesting judgement, and it's implications, if you are correct haha.
1 reply →
The word transformative was put there in a time of manual transformative processes, like when you paint something similar to what you saw in a painting by another artist, with all the implied limitations that entails, like the time it took from you to watch that painting, and the time it takes you to create that new painting, nothing to do at all with the way LLMs operate, an honest assessment would have found that the word was meant for a wildly different use case and therefore it required a bigger and more nuanced discussion.
> The word transformative was put there in a time of manual transformative processes, like when you paint something similar to what you saw in a painting by another artist
Do you have any citation that that is how the word "transformation" was understood historically? Because what your suggesting seems to be the opposite of what i've read.
My understanding is even back in the 1800s (e.g. https://en.wikipedia.org/wiki/Folsom_v._Marsh ) your example would not be considered transformative, if your intention was to make a similar painting to serve a similar purpose.
That’s one hell of a headline for a story about Meta winning summary judgement for most of the claims against them. You’d be forgiven for thinking Meta lost this case, going by the headline.
It's not a story about Meta winning (that previous story is linked in the first sentence), but a story expanding on the part where they didn't. The headline is totally appropriate for that piece.
Surely it should be a whole separate copyright case with fines of up to $150,000 per work infringed.
It’s crazy, yet so predictable, that while the system tries to bankrupt individuals for torrenting a single book or movie in this case the excuse “it was just to train an LLM” will fly. Imagine a private individual would argue that in court.
Ironically, the Llama models enable people to fine-tune on their own material. A lot of people are doing exactly this.
Feels more of political purpose as AI is now a "national strategy".
The system wants to destroy creativity and humans. The system wants artists and writers to depend on the patronage of the ultra rich who steal their output.
We are back to feudal society, except that monarchs or their advisers at least had taste as opposed to the Nouveau riche.
Wait. So I can pirate any stuff as long as I intend to train an LLM with it?
I mean it seems clear that Meta did not pirate the content to watch/read it. However, I guess according to the ruling you could pirate anything you want (but no seeding), produce a shitty haiku based on what you pirated and then claim fair use.
No, because the downloading was considered illegal. You can use any book or movie you bought to train an LLM though.
my brother in Christ, you can pirate any stuff for any reason.
Haha. I meant legally. In EU it's common to get fined for pirating movies.
Sorta not really. They said the plaintiff had a non relevant argument or something.
So the argument is that by torrenting ebooks, meta provided bandwidth to the torrent network, and thus provided (financial??!) benefit to pirate sites?
I got to be honest, that sounds extremely weak to me. The benefit to the pirate site of joining the torrent swam seems like it would be extremely slight.
The pirate sites? There is no pirate site hosting files here, BitTorrent is peer to peer.
If it’s fine for a large corp. like Meta to pirate books then it’s fine for everyone else. If it’s a crime for ordinary consumers then it’s a crime for Meta too.
Especially as Meta aren’t doing this for charity. They train LLMs for their own gain.
> The pirate sites? There is no pirate site hosting files here, BitTorrent is peer to peer.
Or "shadow library" or whatever you want to call it. The argument according to the article is that the entity that created the torrent, which also as far as i understand also operates a traditional website, benefits from meta's actions.
I think that is really far fetched.
The argument is that distribution is the infringement. This is basically the only thing torrenters get charged with.
That does not seem to be the argument that was presented in the article.
When do the outputs cease to be a derived work?
When you're rich.
[dead]