Academic and entertainment publishers say Meta “engaged in one of the most massive infringements of copyrighted materials in history” in a new lawsuit filed on Tuesday in a US District Court in New York.
The claims are familiar: Publishers, including McGraw-Hill, Elsevier, Cengage, Hachette and Macmillan, allege that Meta illegally acquired, or pirated, copies of their copyright-protected materials — scientific journal articles, textbooks and other books — to train its Llama AI models. Author, lawyer and former Authors Guild President Scott Turow is also joining the publishers in the lawsuit.
Meta CEO Mark Zuckerberg is specifically named as a defendant, with the complaint saying the CEO “personally authorized and actively encouraged” the alleged illegal behavior. As a result, Meta’s AI “readily generates, at speed and scale, substitutes for [authors] works on which it was trained.”
“Meta chose to live by its motto of ‘move fast, and break things,’ and now must be held accountable for what it broke, including the copyright laws,” the American Association of Publishers said in a statement. An attorney for the plaintiffs did not immediately respond to a request for comment.
A Meta spokesperson told CNET: “Courts have rightly found that training AI on copyrighted material can qualify as fair use. We will fight this lawsuit aggressively.”
New lawsuit, same questions
Copyright is one of the most contentious legal issues around AI. Tech companies like Meta need high-quality, human-created data to build and refine their AI models. Nearly all of this material is protected by copyright. That means tech companies have to enter into licensing agreements or defend their use of the content as fair use under a provision of copyright law.
Meta and Anthropic have both won previous cases in lawsuits brought by authors, successfully defending their fair use. Anthropic agreed to settle some piracy claims with authors for $1.5 billion, or about $3,000 per pirated work. Both judges warned in their decisions that this won’t be the result in every lawsuit.
US District Court Judge Vince Chhabria wrote in his 2025 ruling for Meta, “The market for the typical human-created romance or spy novel could be diminished substantially by the proliferation of similar AI-created works.”
One of the biggest considerations in these cases is whether tech companies’ use of copyrighted books will make it harder for human authors to sell their work or otherwise affect the marketplace. The plaintiffs argue Meta’s AI models can pop out entirely AI-generated scientific articles and novels, pointing to a number of authors selling AI-written works on Amazon. This is especially concerning for authors who say people are using AI to create content in their specific style.
“I find it distressing and infuriating that one of the top-10 richest corporations in the world knowingly used pirated copies of my books, and thousands of other authors, to train Llama, which can and has produced competing material, including works supposedly in my style,” Turow told The New York Times.
Precedent — the history of prior court rulings — always plays a role in how current lawsuits unfold. But it’s too soon to tell whether this case will play out differently from previous cases in which judges sided with tech companies.
