In a landmark ruling, federal judge William Alsup ruled that Anthropic's use of published books to train its artificial intelligence model without authors' permission is legal. This is the first time a court has recognized an AI company's argument based on the fair use doctrine, setting an important precedent for the tech industry in copyright disputes.
Far-Reaching Industry Impact
This ruling has significant implications for the ongoing AI copyright battle. Writers, artists, and publishers have filed dozens of similar lawsuits against tech giants such as OpenAI, Meta, Midjourney, and Google. While other judges may not necessarily follow Judge Alsup's decision, it provides a legal foundation for courts to lean toward supporting tech companies rather than creative professionals.
The core of the dispute lies in the interpretation of the fair use doctrine—a copyright law provision that has not been updated since 1976, long before the internet and generative AI emerged. Fair use determinations typically consider factors such as the purpose of the use, its commercial nature, and the degree of transformation in the derivative work.
Piracy Issues Remain Under Review
Although Anthropic won the battle over training use, it still faces serious charges. In the case of Bats v. Anthropic, the plaintiffs questioned the way the company obtained the works. The lawsuit claims that Anthropic attempted to create a "central library" containing "all books in the world" and "permanently preserve" them, but millions of copyrighted books were downloaded free of charge from piracy websites.
Judge Alsup clearly stated in his ruling: "We will address the pirated books used by Anthropic to create the central library and the damages they caused. Anthropic later purchased books that had previously been stolen online, which does not absolve it of theft liability, but it may affect the amount of statutory damages."
