The judge said that one «potentially winning argument» — that AI tools could harm the market for human-created content — was barely mentioned.
Meta just scored a major legal win in the battle over how AI models are trained — but not because the court cleared its actions.
Instead, the judge said the lawyers for the other side botched the case.
A California federal judge on Wednesday dismissed most of a lawsuit brought by a group of authors who accused Meta of using their copyrighted books to train its AI models.
The ruling said Meta had used LibGen, a shadow library that hosts millions of pirated books, academic articles, and comics, to train its large language models, including Llama.
Companies like Meta require vast amounts of input to develop their large language models, so they’ve tapped sources from social media posts to videos to books. Authors, artists, publishers, and other groups contend that the use of their work for training amounts to theft.
US District Judge Vince Chhabria said he ruled for Meta not because its use of copyrighted materials is lawful, but because the plaintiffs bungled the argument.
«This ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful», Chhabria said.
