San Francisco: Meta has won a US court case involving AI training and copyright, after a judge ruled that a group of authors failed to prove the company’s use of their copyrighted books to train its AI system was unlawful.
The lawsuit, filed by a group of authors including Sarah Silverman and Ta-Nehisi Coates, alleged that Meta had used pirated versions of their books without consent to train its AI system. This ruling marks a significant development in the ongoing debate over AI training and copyright practices in the tech industry.
The focus of the AI training and copyright case was whether Meta’s training of its LLaMA AI model on copyrighted material constituted infringement. The court determined that the plaintiffs did not present sufficient evidence that the AI tool harmed the market for their books, which is required to prove a violation under US copyright law.
Judge Vince Chhabria emphasized that his ruling did not mean Meta’s actions were entirely lawful, but rather that the authors failed to make the right legal arguments. Judge Chhabria said that training AI with copyrighted works could be illegal in many circumstances, noting that, “These plaintiffs made the wrong arguments and failed to develop a record in support of the right one.”
Meta welcomed the ruling and reaffirmed its support for the fair use doctrine, calling it a vital legal principle that supports innovation in AI development. Meanwhile, the authors’ legal team expressed disagreement, arguing Meta engaged in ‘unprecedented pirating of copyrighted works.’
The decision comes just days after another judge ruled in favor of Anthropic, another AI company, in a similar case. However, Anthropic still faces further trial for allegedly copying over 7 million pirated books.
The AI training and copyright case underscores the broader debate over AI and copyright, as companies argue for fair use while creators fear threats to their livelihoods. Judge Chhabria highlighted that generative AI can flood markets with content produced in far less time than traditional creative processes, potentially undermining the incentive to create.
This ruling adds to the growing debate over AI training and copyright, highlighting legal uncertainty around AI development.
ALSO READ | NASA gears up for Axiom Mission 4 launch to ISS
