Meta has won a major legal battle in a copyright lawsuit filed by 13 authors who claimed the company used their books without permission to train its artificial intelligence systems. The case centred around whether it was legal for the company to copy copyrighted works for training large language models (LLMs).
On Wednesday, Judge Vince Chhabria ruled in Meta’s favour, stating that the company is “entitled to summary judgment on its fair use defence to the claim that copying these plaintiffs’ books for use as LLM training data was infringement.”
However, the judge made it clear that this ruling doesn’t give Meta a free pass to use copyrighted content for AI training. “This ruling does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful,” Judge Chhabria explained, according to The Verge. “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.”
Also read: Meta faces backlash as technical error bans thousands of Facebook groups
In the lawsuit, the authors made several claims about how Meta’s AI model used their work. They argued that Llama could reproduce sections of their books and that Meta’s use of their work harmed their ability to license it for training purposes. But the judge dismissed these points as weak, calling them “clear losers.”
The judge stated, “Llama is not capable of generating enough text from the plaintiffs’ books to matter, and the plaintiffs are not entitled to the market for licensing their works as AI training data.”
Also read: Microsoft plans to cut more jobs at Xbox division next week in latest round of layoffs
According to Judge Chhabria, plaintiffs failed to present a strong enough case to support the claim that Meta’s copying would lead to “a product that will likely flood the market with similar works, causing market dilution.”
This ruling comes a day after Anthropic won in a separate case, where a federal judge ruled that using legally purchased copies of books to train its AI models qualifies as fair use.