Comment by jaggederest
1 month ago
Perhaps even more importantly, the current frontier LLM models are self-admittedly the product of enormous quantities of copyright infringement and even less savory inputs, so calling them out for distilling the fruit of that tainted tree reads as highly hypocritical at best.
No, they settled that yesterday, so all is forgotten. Press releases were queued for today so just in the nick of time.
I think that was specifically on the piracy aspect.
So Moonshot is next for a lawsuit or do lawyers only care about Anthropic?
Between the two of them, only one firm is trying to pull up the ladder so that no one else can have what it took.
This is why I don't give a shit that this is happening. It's actually kind of funny to me.
Unless you’re from mainland China, you should.
Why? Should we be held hostage to the whims of these companies and all the investors in the throes of AI psychosis, and let them do whatever the fuck they want, because if we don't then the economy will crash?
No you should not care unless you are a shareholder in one of these Ai ponzi companies. For the rest of the world Chinese companies matching and open sourcing llm's will keep 100 or so tech oligarchs taking over all the world economic output for themselves as all the idiot politician are unwilling to tax wealth.
3 replies →
Why?
Serious question.
I'm from the US, and I think it's hilarious.
23 replies →
Why so? If I'm from Europe or South America, should I hope Anthropic/Openai win?
6 replies →
Why?
fair use, both in the original training and in distillation, or rather, anthropic has no copyright at all over the output tokens
No - distillation is not data inputs.
Raw materials vs. Value add.
They are different things, like ore and metal.
Distillation is a new thing we need to understand, it's probably closer to IP than not.
The "data inputs" were also, very much, somebody's "value added" IP.
We're talking about things like text people wrote, not some kind of raw data floating out in the ether.
Did I say there was no value add in the inputs?
Ore has value, a different kind of value than the output of the refinery.
distillation has been around for 12 years. it's not new in terms of ML techniques.
https://arxiv.org/abs/1503.02531
although i doubt there has been a legal case over it yet in the context of the legality of stealing shit but IANAL.
Yes, I get that, but it's only now the issues are coming into the commons in a way that industry / society needs the regulatory clarity.
It's completey insane that we still don't know how Open Source would work, that the laws are vague and we're still technically waiting for the courts to decide on cases.
The government should a) legislate and b) create test cases and run them through the courts so that we can have clarity.
58 replies →
Are you suggesting data input is further from IP than distillation?
That would stun me, but it's a little hard to read.
LLM outputs are not copyrightable or rather the user who generated them owns it.
That entirely settles it and there isn’t much else to say about.
If Anthropic feels that other countries are violating their EULA well they are free to stop doing business with them.
The claim these companies make goes much, much further than that.
They claim LLMs "uncopyright" their inputs. So if I take, say, 50 Mickey Mouse comic books, tell ChatGPT to read them and produce 50 "Buster Beagle" comic books that there is ZERO "copyright contamination" and I own those 50 output comics without Disney having any claims on them whatsoever.
Or if I ask ChatGPT to "make a spreadsheet software like Excel, Sheets, Calc, ..." that, again, there is zero copyright claim possible from these people.
It has not been tested, of course.
Do you think writing books (and Wikipedia articles, and stack overflow articles, and github repos, and, and, and, and ...) is not a value add?? What terrible claim.
While I agree on a moral level, I think there is a distinction to be made. Training a SOTA model takes a huge amount of resources and expertise so the people doing the training are adding a lot of value along the way. I think this is much less true for distillation (which is kind of the whole point).
ed: to clarify, I totally agree that a huge chunk of the value in LLMs is coming from the source material. My point was just that training an LLM takes more resources and expertise than distilling from an existing LLM so I don't think the equivalence between training and distilling is entirely justified.
I like this comment because its argument only makes sense if you assume that the entire world's output of books and art did not require a huge amount of resources and expertise to make, nor did it add any value.
It's the most CS-major take ever!
If turning other peoples copyrighted work into a model is transformative enough to be protected then so is distilling that model into a different, better, model.
22 replies →
there is a major god complex here.
MBAs and non technical managers = inept Catbert-type charlatans.
Software engineers, devs, etc = geniuses capable of mastering any domain, innate ability to be right on any topic.
I don't think that's what it's saying at all. It's saying that there's a level of creativity in model creation that isn't present in distillation.
15 replies →
No? They outright say the opposite!
Like look, I'm not a native speaker, sure. But I think when someone says "value add", that means there was value there (which you claim they're rhetorically erasing), and then that was added to. Under no interpretation of this phrase do I get an erasure of prior value.
So certainly, as long as words mean anything, no, they absolutely did not say or suggest what you claim they did, and what you extract a thus unreasonable amount of obnoxious schadenfreude from, while throwing in a cheap insult for funsies at the end.
It's the second time I feel compelled to reach for this just today: https://i.kym-cdn.com/photos/images/original/002/659/979/108...
This is a misrepresentation though.
The LLM output, is not the same as the input - there is value add.
Of course works used as raw inputs to LLMs required work and are reasonably subject to IP concerns - but they are different.
It's possible that the LLM makers 'owe' the content creators that created the content they used to make their products - it's an interesting but separate question.
We could very well end up where content IP is protected, LLM output is not and visa versa with reasonable legal founding, doubtful but plausible.
5 replies →
> raining a SOTA model takes a huge amount of resources and expertise
Writing books, building Wikipedia, and answering questions on online forums takes a lot of resources and expertise that scraping didn't. So at the very least, we're already one rung down the "maybe you should've asked" ladder.
I suspect that, in aggregate, all of the informational output of humanity prior to 2020 has taken more resources to produce than the last few years of LLM research.
I don't know man. This reads like "yeah we stole your grain, but making bread is hard."
It sure is, but it doesn't matter. Whatever position that generates more economic activity is declared legal using some nonsense retconned logic "because we said so".
Why is it less true for distillation? Everyone technically has access to Fable but Moonshot came up with the model. How can you objectively claim one is adding value while the other is not?
If that is the whole point you need to clarify why this is the case on an objective level.
I would say building a comparable model using any means necessary (just like what Anthropic and OAI did) at a lower cost is actually more valuable to soceity and Monshoot is arguably generating more value with less.
Probably not as much effort as writing books and creating art the models were trained on.
The value of LLM's come from replacing what generated its training data.
If the distilled model is cheaper, then it's just LLM's getting LLM'ed.
Still, AFAIK Kimi's architecture (just like that of other LLMs from Chinese labs) is different from those of OpenAI and Anthropic's model in a nontrivial way. So the expertise is still there, and I guess resource use too (although Chinese labs tend to optimize this, thanks to the restrictions they have on GPU use).
EDIT: just wanted to add that resource optimization is usually where the contribution of Chinese labs is, so you shouldn't reaad the above parenthesis as a negative comment.
As an author, that's a genuinely disheartening thing to read.
It took me a year to write a book. It took OpenAI and Anthropic a fraction of a second to ingest it. Do you understand now why I give zero shits if it takes Anthropic a billion to train a model, and Moonshot 10k in API cost to distill it?
> training an LLM takes more resources and expertise than distilling from an existing LLM
This is not automatically true. Training and distillation use the same underlying infra and method and there is no intrinsic differences in between.
They add value on top of other people’s work, often against licensing, and then commercialize this product, ie profiting from making a product out of other people’s IP.
>Training a SOTA model takes a huge amount of resources and expertise so the people doing the training are adding a lot of value along the way.
producing the entire body of human knowledge that Silicon Valley companies absorbed like the Borg did not just take more resources but also a fair amount of blood and sweat, certainly more than the LLM so on that front that comparison also seems entirely justified.
I'm sure it takes a lot of time and resources to plan and pull off an epic heist but it is unusual to see people like Thomas Crown being accused of creating value, as they're usually accused of committing theft.
Yeah, there's a difference. One party spends a bunch of resources doing something illegal and extremely immoral. The other party spends little money doing something legal and morally neutral.
You can argue that reverse engineering anything is as hard if not harder than engineering something. I can’t imagine distillation is any different.
Distillation is objectively easier than training a model from scratch, that's why all these Chinese labs are doing it.
2 replies →
I disagree that LLM models are the product of enormous quantities of copyright infringement.
The recent announcement that AI-assisted research produced a counterexample to the Jacobian conjecture--a long-standing open problem in algebraic geometry--shows the original value AI can create. The result was not copied from a textbook; it emerged from AI learning from existing material, much as a human does, and then applying that knowledge in a new way. If that's a violation of copyright, then a human doing the exact same thing would be a copyright violation too. But it isn't.
If you re-read your comment, you will find that your second paragraph is not evidence for the claim you make in your first paragraph. In fact, your first paragraph is just false.
Let me explain it this way: If it is legal for a human to learn from a book, then disseminate the knowledge, then it is legal for a machine to do so. You may think this is not right because a machine does it at a much larger scale, but if so laws need to be updated. As it stands now there is no law that says if a human does X it is not a copyright violation but if a machine does the same X it is copyright violation.
1 reply →
Yup if a machine kills a human it's not the machine's fault; it's the human's. Humans doing the exact same thing as machines aren't 1:1.
If it is legal for a human to do something then it is legal for a machine to do it too. Are there any counter examples to that?
3 replies →
The didn’t pay for the books.
It’s massive copyright infringement.
The human buys the books.
It's worth keeping in mind the purpose of copyright. It's a pragmatic tool to encourage investment in creative work for the benefit of everybody/consumers. We may be entering a time where there's less need to incentivize people to write books. At least not non-fiction books which are simply a collection of existing knowledge presented in an a way that's suitable for human readers. A lot of the value those authors provided can now be done by AI. Yes, the AI trained on their work, but now that it's here, we don't need new non-fiction authors quite as much as we used to.
I wouldn't want to live in a world where technology or general people's wellbeing was held back by obsolete laws that ended up lingering on just to protect undeserving special people at the expense of the rest of society. Remember guilds for tradesmen? They were also a monopoly given by the government to special people. They had their purpose but nowadays we have different ways to keep tradesmen working effectively like license requirements and insurance.
Just to be clear, I think we do still need copyright, but that we might be in a transition period where it has to be redesigned to adapt to AI.
1 reply →
Did they borrow the book? If I learn from a borrowed book is that copyright infringement?