US judge considers Anthropic's $1.5 billion settlement of authors' lawsuit

A U.S. federal judge in San Francisco delayed final approval of Anthropic's $1.5 billion settlement with authors over alleged misuse of their books to train its AI chatbot Claude, requesting more details on fees and payments. The case, the largest known U.S. copyright settlement involving AI training, follows a 2024 lawsuit where Judge William Alsup ruled Anthropic’s fair use claim was valid but found piracy violations in storing millions of books." "article": "A federal judge in San Francisco postponed final approval of Anthropic’s proposed $1.5 billion settlement with authors who accused the AI company of misusing their books to train its chatbot Claude. U.S. District Judge Araceli Martinez-Olguin asked for further details on lawyers’ fees and payments to lead plaintiffs, delaying a decision after initially greenlit by retired Judge William Alsup in September 2023. The settlement covers over 92% of the 480,000+ works included, though some authors argue it is insufficient or unfairly compensates attorneys while excluding certain copyright owners. Anthropic, backed by Amazon and Alphabet, faced claims in 2024 of using pirated books without permission to train Claude. Alsup ruled in June 2024 that Anthropic’s fair use defense was valid but found the company violated rights by storing 7+ million pirated books in a central library, not all used for AI training. The case is the first major U.S. AI copyright lawsuit to settle, amid dozens of similar claims by authors and news outlets against tech firms. A trial was scheduled for December to determine damages, potentially reaching hundreds of billions. Some authors, including Dave Eggers and Vendela Vida, filed a separate lawsuit in California on May 15, 2024, opting out of the settlement. Ongoing disputes highlight tensions between AI development and copyright protections, with critics arguing settlements may not fully address broader legal risks for tech companies.