Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Thomson Reuters wins first major AI copyright case in the US

wired.com

Post reply on HN
Print view View original

61–70 of 188 posts

Re: Thomson Reuters wins first major AI copyright case in the US

#61

Post by unyttigfjelltol »

Ajedi32 wrote: post #57
JackC wrote: post #23

Here's the full decision, which (like most decisions!) is largely written to be legible to non-lawyers: https://storage.courtlistener.com/recap/gov.uscourts.ded.721... The core story seems to be: Westlaw writes and owns headnotes that help lawyers find legal cases about a particular topic. Ross paid people to translate those headnotes into new text, trained an AI on the translations, and used those to make a model th…

Interestingly, almost the entirety of the judge's opinion seems to be focused on the question of whether the translated notes are subject to copyright. It seems to completely ignore the question of whether training an AI on copyrighted material constitutes making a copy of that work in the first place. Am I missing something? The judge does note that no copyrighted material was distributed to users, because the AI do…

Ross evidently copied and used the text himself. It's like Ross creating an unauthorized volume of West's books, perhaps with a twist.

Obscurity ≠ legal compliance.

Re: Thomson Reuters wins first major AI copyright case in the US

#62

Post by blibble »

rvz wrote: post #33

See. The fair-use excuses that the AI proponents here were trying to hang on to for dear life have fallen flat on this ruling. This is going to be one of many cases in which there will be licensing deals being made out of this to stop AI grifters claiming 'fair use' to try to side-step copyright laws because they are using a gen AI system. OpenAI ended up paying up for the data with Shutterstock and other news source…

once the case law is set, I look forward to suing everyone that's ever trained a model for $300,000 PER WORK each time they ingested my code from GitHub

whoever wrote those indemnity policies is going to regret it

Re: Thomson Reuters wins first major AI copyright case in the US

#63

Post by mvdtnz »

echelon wrote: post #39
JackC wrote: post #23

Here's the full decision, which (like most decisions!) is largely written to be legible to non-lawyers: https://storage.courtlistener.com/recap/gov.uscourts.ded.721... The core story seems to be: Westlaw writes and owns headnotes that help lawyers find legal cases about a particular topic. Ross paid people to translate those headnotes into new text, trained an AI on the translations, and used those to make a model th…

If the copyright holders win, the model giants will just license. This effectively kills open source, which can't afford to license and won't be able to sublicense training data. This is very bad for democratized access to and development of AI. The giants will probably want this. The giants were already purchasing legacy media content enterprises (Amazon and MGM, etc.), so this will probably further consolidation an…

Open source model builders are no more entitled to rip off content owners than anyone else. I couldn't possibly care any less if this impacts "democratized access" to bullshit generators. At least if the big boys license the content then the rightful owners get paid (and have the option to opt out).

Re: Thomson Reuters wins first major AI copyright case in the US

#64

Post by echoangle »

currymj wrote: post #59
echoangle wrote: post #45

Earlier quoted context omitted.

How is the system inherently generative?

Generative is a technical term, meaning that a system models a full joint probability distribution. For example, a classifier is a generative model if it models p(example, label) -- which is sufficient to also calculate p(label | example) if you want -- rather than just modeling p(label | example) alone. Similar example in translation: a generative translation model would model p(french sentence, english sentence) --…

Do you have some kind of dictionary where I can find this definition? Because I don’t really understand how that can be the deciding factor of „generative“, and the wiki page for „generative AI“ also seems to use the generic „AI that creates new stuff“ meaning.

By your definition, basically every classifier with 2 inputs would be generative. If I have a classifier for the MNIST dataset and my inputs are the pixels of the image, does that make the classifier generative because the inputs aren’t independent from each other?

Re: Thomson Reuters wins first major AI copyright case in the US

#65

Post by mvdtnz »

dkjaudyeqooe wrote: post #44
echelon wrote: post #39

Earlier quoted context omitted.

If the copyright holders win, the model giants will just license. This effectively kills open source, which can't afford to license and won't be able to sublicense training data. This is very bad for democratized access to and development of AI. The giants will probably want this. The giants were already purchasing legacy media content enterprises (Amazon and MGM, etc.), so this will probably further consolidation an…

License what? Every available copyrighted work? Even getting a tiny fraction is not practical. To the contrary, this just means companies can't make money from these models. Those using models for research and personal use wouldn't be infringing under the fair use tests.

> License what? Every available copyrighted work? Even getting a tiny fraction is not practical.

Oh no. Anyway.

Re: Thomson Reuters wins first major AI copyright case in the US

#66

Post by zozbot234 »

Animats wrote: post #29

This isn't really about "AI". It's about copying summaries. Google was fined for this in France for copying news headlines into their search results, and now has to pay royalties in the EU. Westlaw is a summarizing and indexing service for court case results. It's been publishing that info in book form since 1872. Ross was trying to compete with Westlaw, but used Westlaw as an input. West's "Key Numbers" are, after a…

The case involves headnotes, not just key numbers. Your links provide examples of such headnotes, which make it very clear that a lot of human creativity and judgment is involved in authoring them - they're not a matter of purely factual information, such as a phonebook. Thus, the headnotes are copywritten, and translating them to a different language doesn't negate that copyright. This looks like a slam dunk case, but it has very little to do with AI training as such - the AI was only used to create a kind of rough indexing over the translated text.

If this was only about key numbers, it might have gone the other way because the fact-like element there is considerably greater.

Re: Thomson Reuters wins first major AI copyright case in the US

#67

Post by currymj »

echoangle wrote: post #64
currymj wrote: post #59

Earlier quoted context omitted.

Generative is a technical term, meaning that a system models a full joint probability distribution. For example, a classifier is a generative model if it models p(example, label) -- which is sufficient to also calculate p(label | example) if you want -- rather than just modeling p(label | example) alone. Similar example in translation: a generative translation model would model p(french sentence, english sentence) --…

Do you have some kind of dictionary where I can find this definition? Because I don’t really understand how that can be the deciding factor of „generative“, and the wiki page for „generative AI“ also seems to use the generic „AI that creates new stuff“ meaning. By your definition, basically every classifier with 2 inputs would be generative. If I have a classifier for the MNIST dataset and my inputs are the pixels of…

if you have an MNIST classifier that just takes in images, and spits out a probability of digits 1-9, that wouldn't necessarily be generative, if it is only capable of modeling P(which digit | all pixels).

But many other types of model would give you a joint distribution P(which digit, all pixels), so would be generative. Even if you only used it for classification.

https://en.wikipedia.org/wiki/Generative_model

I guess these days "generative" must mean "it is used to generate outputs that look like the training data".

But until recently, the meaning had to do with the information in the model, not how it's used.

Re: Thomson Reuters wins first major AI copyright case in the US

#68

Post by zozbot234 »

currymj wrote: post #67
echoangle wrote: post #64

Earlier quoted context omitted.

Do you have some kind of dictionary where I can find this definition? Because I don’t really understand how that can be the deciding factor of „generative“, and the wiki page for „generative AI“ also seems to use the generic „AI that creates new stuff“ meaning. By your definition, basically every classifier with 2 inputs would be generative. If I have a classifier for the MNIST dataset and my inputs are the pixels of…

if you have an MNIST classifier that just takes in images, and spits out a probability of digits 1-9, that wouldn't necessarily be generative, if it is only capable of modeling P(which digit | all pixels). But many other types of model would give you a joint distribution P(which digit, all pixels), so would be generative. Even if you only used it for classification. https://en.wikipedia.org/wiki/Generative_model I gu…

You can derive the latter information (the joint distribution), given the former and a prior over "all pixels"-like data. So, the defining feature of "generative" models is that they feature a prior over their input data?

Re: Thomson Reuters wins first major AI copyright case in the US

#69

Post by anon373839 »

JackC wrote: post #23

Here's the full decision, which (like most decisions!) is largely written to be legible to non-lawyers: https://storage.courtlistener.com/recap/gov.uscourts.ded.721... The core story seems to be: Westlaw writes and owns headnotes that help lawyers find legal cases about a particular topic. Ross paid people to translate those headnotes into new text, trained an AI on the translations, and used those to make a model th…

This is an interesting opinion, but there are aspects of it that I doubt will stand the test of time.

One aspect is the court’s ruling that West’s headnotes are copyrightable even when they merely quote a court opinion verbatim, because the editorial decision to quote the material itself shows a “creative spark”. It really isn’t workable — in law specifically - for copyright to attach to the mere selection of a quote from a case to represent that case’s holding on an issue. After all, we would expect many lawyers analyzing the case independently to converge on the same quotes!

The key fact underlying all of this, I think, is that when Ross paid human annotators to write their own versions of the headnotes, they really did crib from West’s wholesale rather than doing their own independent analysis. Source text was paraphrased using curiously similar language to West’s paraphrasing. That, plus the fact that Ross was a directly competing product, is what I see as really driving this decision.

The case has very little to say about the more commonly posed question of whether copyright is infringed in large-scale language modeling.

Re: Thomson Reuters wins first major AI copyright case in the US

#70

Post by currymj »

zozbot234 wrote: post #68
currymj wrote: post #67

Earlier quoted context omitted.

if you have an MNIST classifier that just takes in images, and spits out a probability of digits 1-9, that wouldn't necessarily be generative, if it is only capable of modeling P(which digit | all pixels). But many other types of model would give you a joint distribution P(which digit, all pixels), so would be generative. Even if you only used it for classification. https://en.wikipedia.org/wiki/Generative_model I gu…

You can derive the latter information (the joint distribution), given the former and a prior over "all pixels"-like data. So, the defining feature of "generative" models is that they feature a prior over their input data?

Yes, though maybe not explicitly written down.
Post reply on HN