Federal Judge Rules in Favor of Anthropic in AI Copyright Case

In a major development for the artificial intelligence industry, a federal judge has ruled that Anthropic, a startup backed by Amazon, did not infringe on authors’ copyrights by using books to train its AI model, Claude. The court determined that the company’s use of copyrighted materials qualified as fair use and was considered transformative under existing copyright law.

The case, filed in the U.S. District Court for the Northern District of California, was brought by authors Andrea Bartz, Charles Graeber, and Kirk Wallace Johnson. They alleged that Anthropic built its business on the unauthorized use of hundreds of thousands of copyrighted books. Central to the case was a collection of around 7 million books that the company had allegedly obtained without permission and stored as part of a digital research library.

Despite the authors’ claims, the court found that Anthropic’s approach to training its large language model did not involve reproducing the original works in a way that would violate copyright law. Instead, the model learned from patterns and structures in the content to generate new text, rather than copying or mimicking the creative style of any individual author. This form of use was found to be sufficiently different from the original purpose of the works, qualifying it as transformative and protected under fair use. 

This decision is a noteworthy victory for companies developing generative AI technologies. It suggests that using existing works to train AI systems, when done in a way that does not directly replicate or publicly distribute the originals, can be legally permissible. It also highlights how courts are beginning to interpret intellectual property laws in the context of modern AI development.

However, the case is far from over. While the court sided with Anthropic on the primary fair use claim, it also ordered a trial to assess how the company handled the allegedly pirated materials. Although Anthropic later purchased legitimate copies of some of the books, the court indicated that this action would not necessarily eliminate liability for the initial use of unauthorized copies. The outcome of the trial will determine whether damages are owed and to what extent.

This ruling marks an important moment in the evolving relationship between copyright law and artificial intelligence. As legal standards begin to take shape, content creators and AI developers alike will need to navigate these issues carefully. For creators, the ruling raises important questions about how their work is used without direct consent. For AI companies, it provides some legal clarity but also emphasizes the importance of responsible data sourcing practices.

At ARS Counsel, we continue to monitor these developments closely. As the legal landscape around AI and intellectual property continues to shift, we’re here to help our clients stay informed and protected, whether they’re building the next breakthrough in technology or seeking to safeguard their creative work. Contact us today for a free consultation!