Anthropic’s $1.5 billion copyright settlement approved by US court in landmark AI case

A U.S. federal court has approved a landmark $1.5 billion copyright settlement between artificial intelligence company Anthropic and thousands of authors and publishers, bringing one of the most closely watched AI copyright disputes to a close. The case centered on allegations that Anthropic used copyrighted books without proper authorization to train its AI chatbot, Claude.

The settlement, the largest of its kind in U.S. copyright history, covers nearly 500,000 literary works, with eligible authors and publishers receiving $3,000 per work. More than 91% of those covered by the agreement have already claimed their payments, while the court also approved over $101 million in legal fees.

The lawsuit, originally filed in 2024, accused Anthropic of obtaining millions of books through both legitimate purchases and unauthorized downloads from online piracy websites. Although the court previously ruled that using copyrighted material for AI training could qualify as fair use, it concluded that acquiring books from illegal sources violated copyright law.

Rather than proceed to trial, Anthropic agreed to settle the case, avoiding the possibility of additional financial penalties. Company representatives said the agreement reflects the court's earlier position that AI model training itself remains protected under existing fair-use principles.

The ruling is expected to influence future legal battles involving AI companies, as courts increasingly examine not only how artificial intelligence systems are trained but also how training data is collected. Several lawsuits against major technology firms, including Google, Meta, OpenAI, and Midjourney, continue to test the legal boundaries of AI development and copyright protection.

The decision marks a significant moment in the rapidly evolving relationship between artificial intelligence, intellectual property rights, and the future of creative industries.