Anthropic Reaches $1.5 Billion Deal to Settle Authors' Class Action Over Pirated Training Data
Anthropic will pay $1.5 billion plus interest to about 500,000 authors, averaging $3,000 per work, settling claims of using pirated books for AI training without licenses.
- Anthropic, an AI firm based in San Francisco, has reached a settlement to resolve a class-action lawsuit filed by authors who claimed their books were used without permission to train the company’s chatbot, Claude, agreeing to pay $1.5 billion as part of the deal.
- The lawsuit arose from Anthropic downloading over seven million digitized books from pirate sites like Library Genesis, leading to a June mixed ruling that found fair use by training but infringement by storing pirated copies, with a trial set for December to determine damages.
- The settlement pertains to about 500,000 books after removing duplicates and non-copyrighted titles, includes a payment of approximately three thousand dollars for each book along with accrued interest, and obligates Anthropic to eliminate the pirated datasets it obtained.
- Authors’ lawyer Justin Nelson called the deal likely the largest copyright recovery ever, and plaintiffs said the settlement sends a strong message to AI companies about consequences for pirating works, though European groups say it offers limited help to their writers.
- If approved by a judge at a Monday hearing, the settlement could set a precedent in AI copyright law amid ongoing lawsuits, while Anthropic continues advancing AI with new funding and a projected $5 billion revenue this year.
324 Articles
324 Articles
Anthropic, Claude's developer, one of the most widely used generational intelligence chatbots (AIs), has reached a historic out-of-court settlement with a group of authors and editors. The company will pay them at least $1.5 billion to avoid having to settle a lawsuit in court for having trained their models with the authors' books obtained without paying copyright.
Anthropic to pay authors $1.5B to settle lawsuit over pirated books
NEW YORK — Artificial intelligence company Anthropic has agreed to pay $1.5 billion to settle a class-action lawsuit by book authors who say the company took pirated copies of their works to train its chatbot. Read more...
Coverage Details
Bias Distribution
- 57% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium