Professors, Publishers Back Authors Suing Meta Over AI Training

Publishing groups and law professors defended authors suing Meta Platforms Inc. over its generative AI training process, in a number of friend-of-the-court briefs.

In separate amicus curiae briefs filed April 11, copyright law professors; the International Association of Scientific, Technical and Medical Publishers; Copyright Alliance; and Association of American Publishers said Meta’s use of thousands of pirated books can’t qualify as fair use under copyright law.

The briefs in the US District Court for the Northern District of California came after Meta’s training process received support from the Electronic Frontier Foundation digital rights group and another group of intellectual property professors in early April.

While they argued the court should look at previous rulings involving large-scale copying, like one about Google Books from the Second Circuit in 2015, the copyright professors’ April 11 filing said that product differed from an AI model in that it told users something “about” the books. The technology wasn’t focused on “exploiting expressive elements of the works,” they said.

“Put differently, the expressiveness of the works was beside the point,” the professors wrote. “The technology at issue was content agnostic.”

The professors added Meta’s use wasn’t transformative because, like the AI models, the plaintiffs’ works also increased “knowledge and skill.” To favor Meta’s fair use argument would have a “cascading effect,” the brief said, because several other companies like OpenAI, Google, Anthropic, xAI are building generative AI models using protected content.

Richard Kadrey, Sarah Silverman, and several other authors sued Meta in July 2023 over the tech giant’s use of books to train its large language models. Their case is one in dozens targeting AI models over AI training on copyrighted books, news articles, music, and more.

The Kadrey case is currently looking at two pending motions for summary judgment: one from the plaintiffs who asked the judge to rule in their favor on a direct copyright infringement claim, and one from Meta which asked the court to rule its use of books was legal under the fair use doctrine.

The International Association of Scientific, Technical and Medical Publishers filed a brief supporting the plaintiffs’ motion, arguing Meta’s use of pirated books from sites like LibGen flout copyright law.

“While Meta attempts to label them ‘publicly available datasets,’ they are only ‘publicly available’ because those perpetuating their existence are breaking the law,” STM wrote.

Using “stolen copies” shouldn’t be “conflated with choosing not to license,” it added.

Tycko & Zavareei LLP represents the copyright law professors. Oppenheim & Zebrak LLP and Coblentz Patch Duffy & Bass LLP represent STM. Pryor Cashman represent the Copyright Alliance. Frankfurt Kurnit Klein & Selz PC represents the AAP.

The case is Kadrey v. Meta Platforms Inc., N.D. Cal., No. 3:23-cv-03417, amicus curiae briefs filed 4/11/25.