DRM bros vs AI bros
It's like watching bullies fight. I want both of them to lose but in the meantime, I'm just gonna just enjoy the popcorn and cheer whenever a punch is thrown
DRM bros vs AI bros
It's like watching bullies fight. I want both of them to lose but in the meantime, I'm just gonna just enjoy the popcorn and cheer whenever a punch is thrown
I don't see how this would go any differently to the book lawsuit they settled.
With that lawsuit, the court found that training the LLM on books is covered under fair use, as long as the books are legally acquired and they do not make illegal copies. (obviously, they can't use data fron Anna's Archive any more)
The argument is that them training on the books does not impact sales of the books themselves - someone that was going to buy the book would still buy it even if an LLM was trained on it, just like they'd still buy it even if summaries/CliffNotes and reviews exist. The training is considered transformative as the LLM learns the relationship between the words rather than retaining the text of the books as-is, and transformative use is allowed under fair use.
This is why the AI companies are destroying books - they have to destroy any books they scan in order to comply with US copyright law. Transforming the book from one form to another (eg from print to digital) is OK as long as only one copy remains.
I'm not saying I agree or disagree, just that this is the current state of copyright caselaw in the USA.
Having said that, the music industry is significantly more litigious than the book industry, so we'll see what happens...
A couple of things I think are slightly off:
I always thought there was in obvious win where companies doing this could be forced to archive the scan publicly (after some period of time). They'd be able to use it/sell it for training, but we wouldn't lose rare books forever.
They destroy the books because it allows them to scan them bette
That's definitely one reason, but the copyright argument is also a part of it. The court explicitly said that their digitization is legal only if does not increase the number of copies of the book.
There's no need to do this legally - they could donate used books. The argument shows that the final product is transformative, so it doesn't matter whether they keep the books or not.
They have to keep the digital copy of the book because they add it to the training corpus for the LLM. Selling or donating the original physical book after doing that would void the fair use argument.
LLMs can reproduce quite long passages of books, though I don't think this changes the argument much, because it's not reliable or useful.
One of the tests that determines if it's fair use or not is whether it can serve as a replacement for the original book. Pirated copies can, which is why they're illegal. A summary like CliffNotes can't. Even if the LLM can reproduce long passages, you can't do that reliably (like you said) and it won't work for all books.
I always thought there was in obvious win where companies doing this could be forced to archive the scan publicly (after some period of time)
I definitely agree with this. I think copyright law needs to be modernized to handle cases like this. I think the AI companies should be allowed to donate the digital copy to a library (like the Internet Archive) while still being allowed to keep their copy in their training corpus.
I think copyright law already needed to be modernised... and the way I think it could be done well is consistent with AI training needs. So the whole thing doesn't really bother me. I'm surprised this angle incenses Lemmy so much because I'd have thought we'd on the whole by very anti-copyright in its current form.
The training is considered transformative as the LLM learns the relationship between the words rather than retaining the text of the books as-is
Which is what humans do when learning something using existing works without paying the artist. People tend to forget that the goal of neural networks was to programmatically mimic the way the human brain works.
Of course the mechanisms behind it are different, but they'd have to be because it's silicon vs carbon based processing.
Shouldn't this be posted in UpliftingNews?
When they win
Neh... they're all asshats, too. It's the artists that need to win
If you want to support artists, go to concerts and buy merch. They make almost nothing on music sales/plays.
I know... and I do.
You sill can't steal music. That's not how it works.
This may be unauthorised use, derivative works, possibly copying, but theft is the wrong word and property doesn't apply to works of art, only to the physical medium they are etched on.
Let’s play the oligarch monopoly illegal antitrust violating merger game… CBS and Paramount is skydance which wants to merge Warner, all run by David Ellison, son of Larry Ellison, which is Oracle, which works with OpenAI, so it totally tracks they would want to sue Anthropic.
This strategy is super familiar. This is almost like how Oracle acquired Sun Microsystems, to obtain Java to sue Google over Android… Hmmmmmm
This might be the first time I'm actually rooting for Anthropic. I mean I do use Claude mostly because it's cheap as hell on a subscription and Kimi has pretty nasty terms of use... But I don't LIKE the company. But if it's them vs the Zionellison family, they better win.
Fucking finally! Now let them fight.
I refuse to hope for a world where copyright and AI are both destroyed in one lawsuit because it’s too beautiful to ever be real.
I'm concerned how AI often leads to people to defend copyright. I fear it will come out of the fight stronger than ever, while not slowing AI in any meaningful way.
Oh, I'm not defending copyright- I just think that everyone should play by the same rules.
I agree, the big corporations, their CEOs, their shareholders, and their employees should follow in the exact same footsteps as Aaron Swartz. All of them.
For those that don't know, Aaron was a hero.
The result is going to be some kind of deal where AI use requires to pay a certain percentage or something like $0.01 per token in copyright fees. This will be the end of open weight models you're allowed to run locally on your PC. It'll be "piracy".
It will strengthen the AI corporations by allowing complete monopolization of AI.
Kind of an either/or deal here. If AI loses, copyright wins (and it's David fucking Ellison, son of Oracle himself). If copyright loses, AI wins.
Just latching ointo the top comment: this page compares AI-generated music with the originals it blatantly stole from. It's against Suno, not Anthropic, but it illustrates the overall issue pretty well.
Obligatory (for Germans): GEMA is not necessarily the good guy here. Neither are Sony and Warner.
the oligarchs will settle between themselves on a way to fuck us over that's ok for both of them, as always.
Fucking crazy thing is that there have always been big ass people in the same room watching their shit get stolen. Warner, Disney, etc...
I can't imagine the legal pain if all of the media entities stopped paying ball.
Uh AI steals everyone's work. How it's legal I'll never know.
because it was ruled to be a transformative work, like how someone making a video essay on a book/game/movie is (sometimes) not violating copyright.
If AI continues to be treated as transformative work, there should also be a valid case to kill licensing requirements for sampling music. After all, there you're literally using small clips of someone's music or similar to make undeniably new art.
So if I modify someone else's art and call it my own, the courts will respect this precedent and let me go?
I love intellectual property theft.
Mind to explain why? Are you implying everything should be CC0? Are you saying what AI companies are doing is good? Should one license their software as CC0 instead of GPL? (sry, I’m not fully versed in the IP apparatus, so pls ignore my comment if I sound dumb. I will do some more research on it anyway at some point)
I think IP law shouldn't exist, because it assumes that ideas can be owned and leads to predictably absurd results.
can they though? Isn’t the purpose of copyright mainly to manage industries e.g. avoid that one actor passes off their work as someone else’s? And they need to have a expiration date. That’s how I understood it. I can see how it can turn in to a cluster fire easy though, you’re right
And now watch everyone cheer for Sony and Warner to succeed in their lawsuits to extend the grip of copyright. What a timeline.
Wonder if Larry Ellison being a direct line between the US presidency and Israeli intelligence would change anything about people's opinion on Warner winning... Probably not.
Nah, people are not THAT stupid.
A while back I saw a thread in a piracy community that was cheering some other lawsuit against an AI company for similar reasons, so I wouldn't bet. The anti-AI sentiment is quite strong in these parts.
They’re not going to walk back copyright law so I’ll take these rapacious AI companies being held a little closer to the same standard that we are.
They’re not going to walk back copyright law
Why not? These laws exist the make the rich more money; if the interests of the rich people trying to make money change it's very possible for the law to change to match that.
Because there’s more money to be made in keeping it than walking it back. I could see them making some kind of shitty carve out for the AI companies that they have to pay into I guess but nothing that would benefit the average person.
We are talking about "Property" - the core foundation of Capitalism. The holy grail of exploitation. Everything in this manufactured belief-system must have an owner and a price, and that price can only be forced upon others with 'property-rights' + top-down force (acab's).
So, just about everything in our sick western fascist/oligarch societies is centered around 'Property' and the ideological right and moral duty to maximise profits from that property -> extortion.
The balance might shift slightly/temporarily if a huge cnt are fighting other huge cnt's, but will NOT go away as long as Capitalism is the sole driver of western societies. Maximum property-rights are the steaming turd every Capitalist cnt are eating/craving.
More infighting is imho better as it slows down this pscho ideology..
This isn't a question of "walking it back", though. There was never anything in copyright law that prohibited analyzing a copyrighted work before. This is an attempt to extend it.
Meta literally torrented 82TB of pirated material off anna’s archive and similar sources they’re not just “analyzing” things they paid for. Not to mention the various terms of use stating that content can’t be used to train off of that they violate as they scrape the whole web.
And so, exactly as I predicted, we've got commenters coming out to defend the poor innocent copyright cartels against those awful people pirating their stuff. Violating sacred EULAs, even, what monsters.
If I could get rid of both of them I would but I can’t so I don’t think it’s unreasonable have them go after the companies burning the planet to make the worst slop imaginable instead of people writing switch emulators or whatever. “Violating those sacred EULAs” causes the very site I’m posting this from to fight a constant game of cat and mouse with the bots trying to scrape it.
That's not how this works though, hoping the law to just punishes people you don't like without regard for what they are being punished for is madness.
If Anthropic lose this they'll shrug pay a fee to licence things out of the untold billions they have available to them from credulous investors, and everyone else will have lost the right to do whatever they want with a copyrighted work without the publisher's permission. That people who consider themselves progressive think that's a win boggles my mind.
We already don’t have the right to do whatever we want with a copyrighted work without the publishers permission.
all 16 comments