Meta has secured a pivotal legal victory in a copyright lawsuit involving 13 authors who accused the tech giant of unlawfully training its artificial intelligence systems on their works without consent. On Wednesday, Judge Vince Chhabria ruled in favor of Meta, asserting that the company is “entitled to summary judgment on its fair use defense regarding the claim that copying these plaintiffs’ books for use as LLM training data constituted infringement.”
Nevertheless, Judge Chhabria highlighted some significant weaknesses in both Big Tech’s AI initiatives and Meta’s fair use defense. He clarified, “This ruling does not imply that Meta’s use of copyrighted materials to train its language models is lawful.”
He further explained, “This ruling merely indicates that these plaintiffs presented inadequate arguments and failed to compile sufficient evidence for a stronger case.” This decision follows a recent fair use victory for Anthropic in a separate ruling, which concluded that training its models on legally obtained copies of books constitutes fair use.
Judge Chhabria identified two arguments raised by the authors regarding fair use that he deemed “clear losers.” These involved Meta’s Llama AI generating snippets of text from their books and the claim that Meta’s usage diluted the market for licensing those works for AI training. He noted, “Llama is not capable of producing enough text from the plaintiffs’ books to be significant, and the plaintiffs cannot claim the market for licensing their works as AI training data.”
According to Judge Chhabria, the plaintiffs also failed to make a convincing case that Meta’s actions would generate “a product that would likely inundate the market with similar works, leading to market dilution.” He referenced the ruling by Judge William Alsup in the Anthropic case, noting that concerns about the potential harm generative AI might unleash on the market for the original works it is trained on were dismissed.
