Judge reviews $1.5B Anthropic settlement proposal with authors over pirated books for AI training

A federal judge has begun reviewing a landmark class-action settlement agreement between the artificial intelligence company Anthropic and book authors who say the company took pirated copies of their works to train its chatbot.

The company has agreed to pay authors and publishers $1.5 billion, amounting to about $3,000 for each of an estimated 500,000 books covered by the settlement.

But U.S. District Judge William Alsup has raised some questions about the details of the agreement and asked representatives of author and publisher groups to appear in court Monday to discuss.

A trio of authors—thriller novelist Andrea Bartz and nonfiction writers Charles Graeber and Kirk Wallace Johnson—sued last year and now represent a broader group of writers and publishers whose books Anthropic downloaded to train its chatbot Claude.

Johnson, author of “The Feather Thief” and other books, said he planned to attend the hearing on Monday and described the settlement as the “beginning of a fight on behalf of humans that don’t believe we have to sacrifice everything on the altar of AI.”

Alsup dealt the case a mixed ruling in June, finding that training AI chatbots on copyrighted books wasn’t illegal but that Anthropic wrongfully acquired millions of books through pirate websites. Had Anthropic and the authors not agreed to settle, the case would have gone to trial in December.

© 2025 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed without permission.