Court rulings leave AI training on copyrighted books in legal limbo

Key Points

• AI models are trained on massive datasets that include copyrighted books and articles.
• Judge William Alsup ordered Anthropic to pay $1.5 billion for pirating texts but upheld the legality of its training methods.
• Attorney Cathy Gellis says the ruling favors AI companies by distinguishing reading from copying.
• Fair‑use analysis hinges on whether AI training is considered transformative, as seen in the Reuters v. Ross Intelligence case.
• The Thaler v. Perlmutter decision declared fully AI‑generated works ineligible for copyright protection.
• U.S. copyright law, unchanged since 1976, forces judges to apply old statutes to new technology.
• Legal experts warn that the lack of clear precedent creates uncertainty for both authors and AI developers.