Kadrey v. Meta Platforms was a copyright suit brought by a group of thirteen authors - including Richard Kadrey, comedian Sarah Silverman, and novelist Ta-Nehisi Coates - alleging that Meta used their books, many obtained from pirated repositories such as the Books3 dataset and shadow libraries, to train its Llama family of large language models. The case was filed in the US District Court for the Northern District of California.
On June 25, 2025, Judge Vince Chhabria granted summary judgment to Meta on the authors’ core copyright claim, holding that training Llama on the books was fair use because the use was “highly transformative.” But the opinion was a warning shot dressed as a win. Chhabria stressed that his ruling “does not stand for the proposition that Meta’s use of copyrighted materials to train its language models is lawful. 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.” He pointed to “market dilution” - the idea that AI-generated works could flood and depress the market for human-authored books - as the argument that could decide future cases, and suggested it “will often cause plaintiffs to decisively win” the market-harm factor.
The decision landed just two days after Judge William Alsup’s ruling in Bartz v. Anthropic, which also found training fair use but treated pirated copies differently. Together, the two San Francisco rulings created an early, partly conflicting roadmap for how courts weigh the four fair-use factors against AI training.
Why business readers should care: Kadrey shows that even a fair-use victory can be fragile. The judge effectively told future plaintiffs how to win, putting AI developers on notice that the “market dilution” theory remains a live and serious threat.