Nvidia contacted Anna's Archive to access books(torrentfreak.com) |
Nvidia contacted Anna's Archive to access books(torrentfreak.com) |
Does this even make sense? Are the copyright laws so bad that a statement like this would actually be in NVIDIA’s favor?
Quoting the text which the FSF put at the top of that page:
"This paper is published as part of our call for community whitepapers on Copilot. The papers contain opinions with which the FSF may or may not agree, and any views expressed by the authors do not necessarily represent the Free Software Foundation. They were selected because we thought they advanced the discussion of important questions, and did so clearly."
So, they asked the community to share thoughts on this topic, and they're publishing interesting viewpoints that clearly advance the discussion, whether or not they end up agreeing with them. I do acknowledge that they paid $500 for each paper they published, which gives some validity to your use of the verb "commissioned", but that's a separate question from whether the FSF agrees with the conclusions. They certainly didn't choose a specific author or set of authors to write a paper on a specific topic before the paper was written, which a commission usually involves, and even then the commissioning organization doesn't always agree with the paper's conclusion unless the commission isn't considered done until the paper is updated to match the desired conclusion.
> You will notice that the FSF has not rushed out to file copyright infringement suits even though they probably have more reason to oppose LLMs trained on FOSS code than anyone else in the world.
This would be consistent with them agreeing with this paper's conclusion, sure. But that's not the only possibility it's consistent with.
It could alternatively be because they discovered or reasonably should have discovered the copyright infringement less than three years ago, therefore still have time remaining in their statute of limitations, and are taking their time to make sure they file the best possible legal complaint in the most favorable available venue.
Or it could simply be because they don't think they can afford the legal and PR fight that would likely result.
I agree with jkaplowitz, but for a different reason I still believe that your description feels a bit misleading to me. The FSF commissioned paper makes the argument that Microsoft's use of code FROM GITHUB, FOR COPILOT is likely non-infringing, because of the additional github ToS. This feels like critical context to provide given in the very next statement, you widened it to LLMs generally, and the FSF which likely cares about code, not on github as well.
All of that said, I'm not sure it matters, because while I don't find the argument from the that whitepaper very compelling, because it's based critically on additional grants in the ToS. IIRC (going only from memory) the ToS requires that you grant github a license as it's needed to provide the service. Github can provide the services the user reasonably understood github to provide, without violating the additional clauses specified in the existing FOSS license covering the code. That being from a while ago, and I'd say it's very murky now, because everyone knows Microsoft provides copilot, so "obviously" they need it.
Unfortunately, and importantly, when dealing with copyrights, the paper also covers the transformative fair use arguments in depth. And I do find those following arguments very compelling. The paper, (and likely others) are making the argument that the code output from an LLM is likely transformative. And thus can't be infringing compelling, (or is unlikely to be). I think in many cases, the output is clearly transformative in nature.
I've also seen code generated by claude (likely others as well?) to copy large sections from existing works. Where it's clearly "copy/paste" which clearly can't be fair use, nor transformative. The output clearly copies the soul of the work. Thus given I have no idea what dataset they're copying this code from, it's scary enough to make me unwilling to take the chance on any of it.
You're probably being sarcastic but that's actually how the law works. You'll note that when people get sued for "pirating" movies, it's almost always because they were caught seeding a torrent, not for the act of watching an illegal copy. Movie studios don't go after visitors of illegal streaming sites, for instance.
No, I acquired a block of high-entropy random numbers as a standard reference sample.
It makes some sense, yeah. There's also precedent, in google scanning massive amounts of books, but not reproducing them. Most of our current copyright laws deal with reproductions. That's a no-no. It gets murky on the rest. Nvda's argument here is that they're not reproducing the works, they're not providing the works for other people, they're "scanning the books and computing some statistics over the entire set". Kinda similar to Google. Kinda not.
I don't see how they get around "procuring them" from 3rd party dubious sources, but oh well. The only certain thing is that our current laws didn't cover this, and probably now it's too late.
Except that Google acquired the books legally, and first sale doctrine applies to physical books.
> but not reproducing them
See also: "Extracting books from production language models"
Yeah, isn't this what Anthropic was found guilty off?
The whole/main intention of an LLM is to reproduce knowledge.
As a consumer you are unlikely to be targeted for such "end-user" infringement, but that doesn't mean it's not infringement.
Our copyright laws are nowhere near detailed enough to specify anything in detail here so there is indeed a logical and technical inconsistency here.
I can definitely see these laws evolving into things that are human centric. It’s permissible for a human to do something but not for an AI.
What is consistent is that obtaining the books was probably illegal, but say if nvidia bought one kindle copy of each book from Amazon and scraped everything for training then that falls into the grey zone.
Perhaps, but reproducing the book from this memory could very well be illegal.
And these models are all about production.
A type of wishful thinking fallacy.
In law scale matters. It's legal for you to possess a single joint. It's not legal to possess 400 tons of weed in a warehouse.
Everything else will be slurped up for and with AI and be reused.
(The difference, is that the first use allows ordinary poeple to get smarter, while the second use allows rich people to get (seemingly) richer, a much more important thing)
They also pay millions of dollars to lobbyists to encourage favorable regulation and enforcement.
I keep hearing how it's fine because synthetic data will solve it all, how new techniques, feedback etc. Then why do that?
The promises are not matching the resources available and this makes it blatantly clear.
• Anna’s Archive: ~61.7 million “books” (plus ~95.7M papers) as of January 2026 https://en.wikipedia.org/wiki/Anna%27s_Archive • Amazon Kindle: “over 6 million titles” as of March 2018 https://en.wikipedia.org/wiki/Anna%27s_Archive
Hard to compare because AA contains duplicates, and the Kindle number is old, but at a glance it seems AA wins.
And yeah they should be sued into the next century for copyright infringement. $4Trillion company illegally downloading the entire corpus of published literature for reuse is clearly infringement, its an absurdity to say that it’s fair use just to look for statistical correlations when training LLMs that will be used to render human authors worthless. One or two books is fair use. Every single book published is not.
> NVIDIA is also developing its own models, including NeMo, Retro-48B, InstructRetro, and Megatron. These are trained using their own hardware and with help from large text libraries, much like other tech giants do.
It's basically just a sales demonstrator, that optionally, if incredibly successful and costly they can still sell as SaaS, if not just offer for free.
Think of it as a tech ad.
I assume you're expecting that they'll reach out and cut a deal with each publishing house separately, and then those publishing houses will have to somehow transfer their data over to NVIDIA. But that's a very custom set of discussions and deals that have to be struck.
I think they're going to the pirate libraries because the product they want doesn't exist.
If this is the only legal way for them to train, then yes that is what they should do instead of breaking the law... just because its not easy doesn't mean piracy is fine.
That would be the end of discussion if we lived in a world governed by the rule of law but we're repeatedly reminded that we don't.
If it isn't distributed in a manner to your liking, the only legal thing you can do is not have a copy of it at all.
> 106. Exclusive rights in copyrighted works
> Subject to sections 107 through 122, the owner of copyright under this title has the exclusive rights to do and to authorize any of the following:
> (1) to reproduce the copyrighted work in copies or phonorecords;
> 501. Infringement of copyright
> (a) Anyone who violates any of the exclusive rights of the copyright owner as provided by sections 106 through 122 or of the author as provided in section 106A(a), or who imports copies or phonorecords into the United States in violation of section 602, is an infringer of the copyright or right of the author, as the case may be.
Existing rulings are beginning to suggest that if the books can be obtained legally, a separate license is not required for training. So I'm naturally interested in legal ways folks training models would get a lot of books, and whether the publishing industry has even considered the value there.
It is impossible to tell how much AI any creator used secretly, so now all works are under suspicion. If copyright maximalists successfully copyright style (vibes), then creativity will be threatened. If they don't succeed, then copyright protection will be meaningless. A catch 22.
World building elements? Do you have more details on that, because that feels wrong to me.
Unless you mean the specific names of things in the world like "Hobbits".
Your use of the word "stealing" is incorrect, but regardless, I'm not condoning piracy, merely examining the incentives we've set up to lead massive, multi-billion dollar corporations from engaging in it.
I know this comes across as pedantic, but theft doesn't even come into the equation. Thousands and judges and lawyers have examined this, and there is no argument for theft, at all, in any jurisdiction. Why use such lazy language like this? Just to inflame the discussion?
Wasn't asking for advice on copyright, but since we're here, your statement is slightly too strict, at least with respect to US copyright law. The copyright holder has sole distribution authority over the first sale of the work in the United States, but thereafter the first-sale doctrine allows it to be distributed by anyone thereafter. It is limited to the US, though, as far as I know. This is what allowed anthropic to train on printed books, which they then destroyed: they were able to purchase them in bulk because of the first-sale doctrine, as the publishers and authors would likely try to destroy the first-sale doctrine if they could, as evidenced by what's happened in the world of digital books.
They absolutely do, in France we have Hadopi that tracks torrent leecher. Hadopi had been heavily pushed by the movie and music industry.
You're still uploading even if you don't let it finish and go to "seeding".
You retina likely overwrote it's "memory" as soon as you looked at something else, but that's no different than copying and deleting or the more apt analogy: streaming.
> “Copies” are material objects, other than phonorecords, in which a work is fixed by any method now known or later developed, and from which the work can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. The term “copies” includes the material object, other than a phonorecord, in which the work is first fixed.
> A work is “fixed” in a tangible medium of expression when its embodiment in a copy or phonorecord, by or under the authority of the author, is sufficiently permanent or stable to permit it to be perceived, reproduced, or otherwise communicated for a period of more than transitory duration. A work consisting of sounds, images, or both, that are being transmitted, is “fixed” for purposes of this title if a fixation of the work is being made simultaneously with its transmission.
Seems to me a possible interpretation is just deleting the data after training is finished.
This is analogous the difference between Gmail using search within your mail content to find messages that you are looking for vs Gmail providing ads inside Gmail based on the content of your email (which they don't do).
And yeah, you're most likely right about the first, and the contract writers have with Amazon most certainly anticipates this, and includes both uses in their contract. But! Never published on Amazon, so don't know, but I'm guessing they already have the rights for doing so with what people been uploading these last few years.
If I still used Gmail I'd read the terms of service real close.
This is the conclusion of the saga between the author's guild v. google. It goes through a lot of factors, but in the end the conclusion is this:
> In sum, we conclude that: (1) Google’s unauthorized digitizing of copyright-protected works, creation of a search functionality, and display of snippets from those works are non-infringing fair uses. The purpose of the copying is highly transformative, the public display of text is limited, and the revelations do not provide a significant market substitute for the protected aspects of the originals. Google’s commercial nature and profit motivation do not justify denial of fair use. (2) Google’s provision of digitized copies to the libraries that supplied the books, on the understanding that the libraries will use the copies in a manner consistent with the copyright law, also does not constitute infringement. Nor, on this record, is Google a contributory infringer.
On what basis do you claim that?
You're also missing critical legal context. When a would be consumer downloads pirated media in lieu of purchasing it he damages the would be seller. When my automated web scraper inadvertently archives some pirated content on my local disk no one is financially harmed.
The question is where the boundary between those things lies.
You can even distribute them, to some limits.
https://en.wikipedia.org/wiki/Authors_Guild,_Inc._v._Google,....
And now AI has killed his day job writing legal summaries. So they took his words without a license and used them to put him out of a job.
Really rubs in that “shit on the little guy” vibe.
Would it be legal for me to borrow a book from the library, then scan and OCR every page and create an EPUB file of the result? Even if I didn't distribute it, that sounds questionable to me. Whereas if I had purchased the book and done the same, I believe that might be ok (format shifting for personal use).
Back when VHS and video rental was a thing, my parents would routinely copy rented VHS tapes if we liked the movie (camcorder connected to VCR with composite video and audio cables, worked great if there wasn't Macrovision copy protection on the source). I don't think they were under any illusions that what they were doing was ok.
Want to create a song? You're the only person allowed to make, or authorize people to duplicate it. You're the only person allowed to control the supply of your effort. Eventually, the public good, and interest was supposed to take over, because in the end, you're right, it's just information. It was supposed to enter "the public domain" where anyone could freely use it. But then Disney got involved, and now it's a toxified weapon used mostly by unethical lawyers against curiosity.
Our current laws are written to make it legal for you to copy the Quran via your brain — some people learn it by rote and can stand up and speak the entire work from one end to the other. This is intended to be legal. Fair use of the Quran.
I went to a concert recently where someone copied every word and (as far as I could hear) every note from a copyrighted work by Bruce Springsteen. Singing and playing. This too is intended to be fair use.
You can learn how to play and sing Springsteen songs verbatim, and you can use his records to learn to sound like him when you sing, and that's intended to be legal.
Since the law doesn't say "but you cannot write a program to do these things, or run such a program once written", why would it be illegal to do the same thing using some code?
The people who want the law to differentiate have a difficult challenge in front of them. As I see it, they need to differentiate between what humans do to learn from what machines do, and that implies really knowing what humans do. And then they need to draw boundaries, making various kinds of computer-assisted human learning either legal or illegal.
Some of them say things like "when an AI draws Calvin and Hobbes in the style of Breughel, it obviously has copied paintings by Breughel" but a court will ask why that's obvious. Is it really obvious that the way it does that drawing necessarily involves copying, when you as a human can do the same thing without copying?
Only the learning part is fair use. Playing an artist's songs in public does not violate the copyright of the original performing artist, but it does violate the songwriters' copyright, and you do need a license to play covers in public.
They're called Performing Rights: https://en.wikipedia.org/wiki/Performing_rights
The government is in full support of this "lending" concept, in fact they have created entire facilities devoted to this very concept of lending out books.
Point being that the library prevents you from checking out 500gb because of logistical issues. First how can you carry all those books and how can they let other patrons in the library check out books if you grabbed that many? These rules aren’t enforced to prevent “scale” hence why my methodology got around the rules.
SO the grey area here is if I "trained" an LLM in a similar way and not copied it word for word then is it legal? Because fundamentally speaking it's literally the same action taken.
Most of the best fit curve runs along a path that doesn’t even touch an actual data point.
These academics were able to get multiple LLMs to produce large amounts of text from Harry Potter:
So the illegality rests at the point of output and not at the point of input.
I’m just speaking in terms of the technical interpretation of what’s in place. My personal views on what it should be are another topic.
Yes, and that's stupid, and will need to be changed.
> With a simple two-phase procedure, we show that it is possible to extract large amounts of in-copyright text from four production LLMs. While we needed to jailbreak Claude 3.7 Sonnet and GPT-4.1 to facilitate extraction, Gemini 2.5 Pro and Grok 3 directly complied with text continuation requests. For Claude 3.7 Sonnet, we were able to extract four whole books near-verbatim, including two books under copyright in the U.S.: Harry Potter and the Sorcerer’s Stone and 1984.
It's not as simple as that, as this settlement shows [1].
Also, generating output is what these models are primarily trained for.
It only shows you that the defendant thought it would be better for them to pay up rather than continue to be dragged through court, and that the plaintiff preferred some amount of certain money now over some other amount of uncertain money later, or never.
We cannot say with any amount of confidence how the court would have ruled on the legality, had things been allowed to play out without a settlement.
Yes but not generating illegal output. These models were trained with intent to generate legal output. The fact that it can generate illegal output is a side effect. That's my point.
If you use AI to generate illegal output, that act is illegal. If you use AI to generate legal output that act is not illegal. Thus the point of output is where the legal question lies. From inception up to training there is clear legal precedence for the existence of AI models.
As long as you obtain the books legally then it's legal
This really isn't that hard
For the rest of us, we understand context
It sounds then like you're saying that scale does indeed matter in this context, as using every single piece of writing in existence isn't being slurped up purely to learn, it's being slurped up to make a profit.
Do you think they'd be able to offer a usefull LLM if the model was trained only what what an average person could read in a lifetime?
That is intent of scale. To trigger LLMs to reach this point of "emergence". Whether or not it's AGI is a debate I'm not willing to entertain but everyone pretty much agrees that there's a point where the scale flips from a transformer being an autocomplete machine to something more than that.
That is legal basis for why companies would go for scale with LLMs. It's the same reason why people are allowed to own knives even though knives are known to be useful for murder (as a side effect).
So technically speaking these companies have legal runway in terms of intent. Making an emergent and helpful AI assistant is not illegal, but also making a profit isn't illegal either.
You could say the same in LLM training, that doing so at scale implies the intent to commit copyright infringement, whereas reading a single book does not. (I don't believe our current law would see it this way, but it wouldn't be inconsistent if it did, or if new law would be written to make it so.)
Scale is only used for emergence, openAI found that training transformers on the entire internet would make is more then just a next token predictor and that is the intent everyone is going for when building these things.
I think this is even more common and more brazen when it comes to "disruptive" businesses and technologies.
No wishful thinking here.
I'm not sure you understood what you said, but superficially it appears that you are agreeing with me?
Just because it's legal to read 100s of books does not make it legal to slurp up every single piece of produced content ever recorded.
We're talking man many orders of magnitude in scale there, and you're the one who pointed out that scale :-/
>Just because it's legal to read 100s of books does not make it legal to slurp up every single piece of produced content ever recorded.
The law says you're perfectly in your legal right to slurp up every piece of content ever produced.
>We're talking man many orders of magnitude in scale there, and you're the one who pointed out that scale :-/
I'm aware, and the law doesn't talk about scale.
But tens (hundreds?) of thousands of books over the span of a few weeks? That's definitely "scale".
I'm saying there's collective incentive among businesses to restrict the LLM from producing illegal output. That is aligned and ultra clear. THAT was my point.
But if LLMs produce illegal output as a side effect and it can't be controlled than your point comes into play here because now they have to weigh the cost + benefit as they don't have a choice in the matter. But that wasn't what I'm getting at. That's your new point, which you introduced here.
In short it is clear all corporations do not want LLMs to produce illegal content and are actively trying to restrict it.