On June 24, a U.S. judge ruled that AI company Anthropic—backed by Amazon and Alphabet—lawfully used copyrighted books to train its Claude large language model under the fair use doctrine.