- cross-posted to:
- technology@lemmy.world
- cross-posted to:
- technology@lemmy.world
Copyright class actions could financially ruin AI industry, trade groups say.
AI industry groups are urging an appeals court to block what they say is the largest copyright class action ever certified. They’ve warned that a single lawsuit raised by three authors over Anthropic’s AI training now threatens to “financially ruin” the entire AI industry if up to 7 million claimants end up joining the litigation and forcing a settlement.
Last week, Anthropic petitioned to appeal the class certification, urging the court to weigh questions that the district court judge, William Alsup, seemingly did not. Alsup allegedly failed to conduct a “rigorous analysis” of the potential class and instead based his judgment on his “50 years” of experience, Anthropic said.
“If we have to pay for the intellectual property that we steal and repackage, our whole business model will be destroyed!”
One thing this whole AI training debacle has done for me: made me completely guilt-free in pirating things. Copyright law has been bullshit since Disney stuck their finger in it and if megacorps can get away with massively violating it, I’m not going to give a shit about violating it myself.
I’m pretty much there too, the whole industry consolidates on the new things and charges us as they make it worse. And there can be some arguments to be made over the benefits of AI but we all know that it will not be immune to the entshitification that has already ruined all the things before it
Same thing but for some reason it’s different. They hate when we use their laws against them. Let’s root they rule against this class action so we can all benefit from copyright being thrown out. Or alternatively it kills AI companies, either way is a win.
YSK. They, we and them in this sentence mean different things to different people.
That’s unfair. They also have to sue people who infringe on “their” IP. You just don’t understand what it’s like to a content creator.
They are very likely to be civilly liable for uploading the books.
That’s largely irrelevant because the judge already ruled that using copyrighted material to train an LLM was fair use.
The judge did so in a summary motion, which means that they have to read all of the evidence in a manner most favorable to the plaintiff and they still decided that there is no way for the plaintiff to succeed in their copyright claim about training LLMs because it was so obviously fair use.
Read the Order, which is Exhibit B to Antrhopic’s appellate brief.
Anthropic admitted that they pirated millions of books like Meta did, in order to create a massive central library for training AI that they permanently retained, and now assert that if they are held responsible for this theft of IP it will destroy the entire AI industry. In other words, it appears that this is common practice in the AI industry to avoid the prohibitive cost of paying for the works they copy. Given that Meta, one of the wealthiest companies in the world, did the same exact thing, it reinforces the understanding that piracy to avoid paying for their libraries is a central component of training AI.
While the lower court did rule that training an LLM on copyrighted material was a fair use, it expressly did not rule that derivative works produced are protected by fair use and preserved the issue for further litigation:
Emphasis added. In other words, Anthropic can still face liability if it’s trained AI produces knockoff works.
Finally, the Court held
Emphasis in original.
So to summarize, Anthropic apparently used the industry standard of piracy to build a massive book library to train it’s LLMs. Plaintiffs did not dispute that training an LLM on a copyrighted work is fair use, but did not have sufficient information to assert that knockoff works were produced by the trained LLMs, and the Court preserved that issue for later litigation if the plaintiffs sought to bring such a claim. Finally, the Court noted that Anthropic built it’s database for training it’s LLMs through massive straight-up piracy. I think my original comment was a fair assessment.
It looks, to me, like you’re reading the briefing without understanding how the legal system functions. You’re making some incredibly basic mistakes. Copyright violations and theft are two distinct legal concepts, for example. You’re treating the case summary as if it were the legal argument in the brief and you’re misinterpreting some pretty clear legal language written by the judge.
No, that is not their argument.
Their legal argument, in the appeal of the class certification, is that the judge did not apply the required analysis in order to certify the three plaintiffs as being part of a class. He instead relied on his intuition, not any discovered facts or evidence. This isn’t allowed when analyzing a case for class certification.
In addition, Anthropic adds, it is well supported in case law (cited in the motion) that copyright claims are a bad fit for class action.
This is because copyright law focuses on individual works and each work has to be examined as to its eligibility for copyright protection, the standing of the plaintiff and if, and how much, of each individual work was the defendant responsible for violating copyright.
This can be done when 3 people claim a copyright violation, because they have a limited set of work which a court can reasonably examine.
A class action would require a court to consider hundreds or thousands of claimants and millions of individual works, each of which can be challenged individually by the defendant.
Courts typically don’t like to take on cases that can require millions of briefings, hearings and rulings. Because of this, courts usually always deny class action certification for copyright violations.
The court, in its order, did not address this or apply any of the required analysis. The class was certified based on vibes, something that doesn’t follow clearly established case law.
This is because training an LLM results in a language model.
A language model is in no way similar to a book and so training one is a transformative use of copyrighted material and protected under fair use.
No, the judge didn’t make any claim about the model’s output after training. That isn’t an issue that’s being addressed in this case. You’re misunderstanding how judges address issues in writing.
Here, the judge is addressing a very narrow issue, specifically the exact claim made by the plaintiff (training with copyrighted material = copyright violation).
The subject of the paragraph is concerned with training the LLM. The claim by the plaintiff is that using copyrighted works to train LLMs is a violation of copyright. That’s what the judge is addressing.
The judge dismissed this argument because it was transformative and so protected by fair use.
The judge further noted that the plaintiffs did not show that training the LLM resulted in “any exact copies nor even infringing knockoffs of their works being provided to the public” and if they could show that training the LLM resulted in “any exact copies nor even infringing knockoffs of their works being provided to the public” then they could bring a case in the future. This is the judge hinting that they can amend their filings in this case to clarify their argument, if they had any evidence to support their claim.
The judge is telling the plaintiff that in order to succeed in their claim, which is that training an LLM on their work is a violation of their copyright, they need to show that the thing that they’re claiming has to result in copies of infringing material or knockoffs.
The training resulted in a model. Creating a model is transformative (a model and a book are two completely different things) and the plaintiffs didn’t show that any infringing works were produced by the training and therefore they have no way of succeeding with their argument that training the model violated their rights.
You’re reading a lot of extra into that statement that isn’t there. The plaintiffs never made a claim about the output of a trained model and so that argument wasn’t examined by the judge.
They’re not stealing anything. Nor are they “repackaging” anything. LLMs don’t work like that.
I know a whole heck of a lot of people hate generative AI with a passion but let’s get real: The reason they hate generative AI isn’t because they trained the models using copyrighted works (which has already been ruled fair use; as long as the works were legitimately purchased). They hate generative AI because of AI slop and the potential for taking jobs away from people who are already having a hard time.
AI Slop sucks! Nobody likes it except the people making money from it. But this is not a new phenomenon! For fuck’s sake: Those of us who have been on the Internet for a while have been dealing with outsourced slop and hidden marketing campaigns/scams since forever.
The only difference is that now—thanks to convenient and cheap LLMs—scammers and shady marketers can generate bullshit at a fraction of the cost and really, really quickly. But at least their grammar is correct now (LOL @ old school Nigerian Prince scams).
It’s humans ruining things for other humans. AI is just a tool that makes it easier and cheaper. Since all the lawsuits and laws in the world cannot stop generative AI at this point, we might as well fix the underlying problems that enable this bullshit. Making big AI companies go away isn’t going to help with these problems.
In fact, it could make things worse! Because the development of AI certainly won’t stop. It will just move to countries with fewer scruples and weaker ethics.
The biggest problem is (mostly unregulated) capitalism. Fix that, and suddenly AI “taking away jobs” ceases to be a problem.
Hopefully, AI will force the world to move toward the Star Trek future. Because generating text and images is just the start.
When machines can do just about everything a human can (and scale up really fast)—even without AGI—there’s no future for capitalism. It just won’t work when there’s no scarcity other than land and energy.
I respectfully disagree. Meta was caught downloading books from Libgen, a piracy site, to “train” it’s models. What AI models do in effect is scan information (i.e., copy), and distill and retain what they view as its essence. They can copy your voice, they can copy your face, and they can copy your distinctive artistic style. The only way they can do that is if the “training” copies and retains a portion of the original works.
Consider Shepard Fairies’ use of the AP’s copyrighted Obama photograph in the production of the iconic “Hope” poster, and the resultant lawsuit. While the suit was ultimately settled, and the issue of “fair use” was a close call given the variation in art work from the original source photograph, the suit easily could have gone against Fairey, so it was smart for him to settle.
Also consider the litigation surrounding the use of music sampling in original hip hop works, which has clearly been held to be copyright infringement.
Accordingly, I think it is very fair to say that (1) AI steals copyrighted works; and (2) repackages the essential portions of those works into new works. Might a re-write of copyright law be in order to embrace this new technology? Sure, but if I’m a actor, or voice actor, author, or other artist and I can no longer earn a living because someone else has taken my work to strip it down to it’s essence to resell cheaply without compensating me, I’m going to be pretty pissed off.
Lol. The liberal utopia of Star Trek is a fantasy. Far more likely is that AI will be exploited by oligarchs to enrich themselves and further impoverish the masses, as they are fervently working towards right now. See, AI isn’t creative, it gives the appearance of being creative by stealing work created by humans and repackaging it. When artists can no longer create art to survive, there will be less material for the AI models to steal, and we’ll be left with soulless AI slop as our de facto creative culture.
That action itself can and should be punished. Yes. But that has nothing to do with AI.
Is that what people think is happening? You don’t even have a layman’s understanding of this technology. At least watch a few videos on the topic.
I think that copying my voice makes this robot a T-1000, and T-1000s are meant to be dunked in lava to save Sarah Connor.
Absurd. It’s their entire fucking business model.
Meaning it would illegal even if they weren’t doing anything with ai…
So what an Ai does is the same thing as every human ever who has read/saw/listened a work and then wrote more words being influenced by that book/artwork/piece.
If you’ve ever done anything artistic in your life, you know that the first step is to look at what others have done. Even subconsciously you will pull from what you’ve seen, heard. To say that AI is not creative because it is derivative is to to say that no human being in history has been creative.
Meta literally torrented an insane amount of training materials illegally, from a company that was sued into the ground and forced to dissolve because of distributing stolen content
“When machines can do just about everything a human can (and scale up really fast)—even without AGI—there’s no future for capitalism.”
This might be one of the dumbest things I’ve ever read.
deleted by creator
Look these people are neck deep in a tribalistic worldview. You cannot reason with anyone who’s against AI simply because their claims are unfalsifiable and contradictory depending on the day and article they are reading. On the one hand it is the shitiest technology ever made which cannot do anything right and at the same time it is an existential threat to humanity.
And I can tell you, that the only reason this is the case is because the right is for AI more strongly than the left. If the right had condemned it, you can be damn right the tables would be turned and everyone who thinks they are left would be praising the tech.
Just move on and take solace in the fact that the technology simply cannot be rebottled, or uninvented. It exists, it is here to stay and literally no one can stop it at this point. And I agree with you, AI is the only tool that can provide true emancipation. It can also enslave. But the fact is that all tools can be used for right or wrong, so this is not inherent to AI.