Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

AI training


AI training might be copyright infringement. But there’s no cases or laws to establish that.

I don’t think this case or anything else has been affected by AI training on copyrighted material, if it is deemed infringing.


It's been demonstrated that some companies, even F10 ones, have been using pirated content to train their AI.


Yes, but not demonstrated that that training is illegal.


why is "training" consumption more legal than "recreational" consumption?

stealing bread to feed the birds vs stealing bread to feed your mom -- both are still stealing


IANAL but my understanding it isn’t about consumption but about distribution and creating derived works. Viewing copyright material isnt illegal, distributing unlicensed copies is. EG, I can loan you a book I bought, or I can loan you a book I stole and you aren’t doing anything illegal in either case.

Stealing bread doesnt matter because stealing physical things deprives the owner of their thing. IP infringement isn’t theft in the legal or moral aspect.


Arguably, it’s worse, because it is commercial use at scale. It’s more akin to public redistribution than private consumption.


https://arstechnica.com/tech-policy/2025/02/meta-torrented-o...

They all seem to be using pirated books. Probably slightly better than just web stuff as it is presumably edited.

The authors case was thrown out on narrow reasoning. But companies now live by different rules so I suspect they won’t be held to account. Even Disney/nintendo are unlikely to stop this…

https://www.pbs.org/newshour/arts/judge-tosses-authors-ai-tr...


What?

Anthropic ($1.5B+ Settlement): In September 2025, Anthropic agreed to pay at least $1.5 billion to settle a class-action lawsuit over using roughly 500,000 copyrighted books from "shadow libraries" to train their Claude LLMs.


> > AI training might be copyright infringement. But there’s no cases or laws to establish that.

> In September 2025, Anthropic agreed to pay at least $1.5 billion to settle a class-action lawsuit over using roughly 500,000 copyrighted books from "shadow libraries" to train their Claude LLMs.

Yes, but not because they were training LLMs with it. The judge in the case found specifically that training the LLMs on the copyrighted material was not copyright infringement; the only copyright infringement Anthropic had committed was acquiring the material itself. In other words, if they had legally bought all of the books they used, they would have been able to train their LLMs on them with no recourse from rights holders.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: