A U.S. federal judge approved Anthropic's $1.5 billion settlement with authors who accused the AI company of using pirated books to train its Claude models. This landmark agreement sets a significant benchmark for copyright risk in the generative AI industry.