Meta’s use of a group of authors’ books to train its artificial intelligence models without permission did not violate the law, a United States federal judge ruled.
The lawsuit was brought by 13 writers, including comedian Sarah Silverman, and alleged Meta had trained its Llama AI models on pirated versions of their books. California district judge Vince Chhabria ruled that these authors did not offer sufficient evidence that Llama’s writing would dilute the market for their work.
“The Court has no choice but to grant summary judgment to Meta on the plaintiffs’ claim that the company violated copyright law by training its models with their books. But in the grand scheme of things, the consequences of this ruling are limited,” Chhabria wrote.
While Meta had claimed that training Llama on these books qualified as fair use, “this ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful,” said Chhabria. “It stands only for the proposition that these plaintiffs made the wrong arguments and failed to develop a record in support of the right one.”
According to Chhabria, the writers’ arguments that Llama could generate parts of their copyrighted texts and diminished their own ability to license their works were “clear losers”. Llama cannot produce large sections of these books and the plaintiffs are not solely entitled to license their works as AI training data, he wrote.
This ruling will only impact the plaintiffs in the case, and does not apply to other authors’ works that Llama may have been trained on without their consent.
In a separate case this week, U.S. district judge William Alsup ruled that Anthropic did not violate the law by training its AI models on pirated copyrighted works, saying that this would be considered fair use. However, Alsup wrote that Anthropic should have paid for access to these books, and could face trial over piracy issues.
Meta also unveiled a US$14.3 billion investment in AI model evaluation company Scale AI earlier this month, and had reportedly weighed buying AI search startup Perplexity AI.
The company’s share price closed at US$708.68, down from its previous close at $712.20. Its market capitalisation is $1.78 trillion.
Related content
