The Anthropic Group has won a significant victory in a legal dispute over copyright and artificial intelligence. This could have repercussions across dozens other AI copyright lawsuits Winding through the US legal system. Anthropic’s AI tool was allowed to be trained on copies of copyrighted work by a court, who argued that such behavior is protected under the law. “fair use” Doctrine that allows the use of certain copyrighted material under specific conditions.
“The training use was a fair use,” senior district judge William Alsup wrote Summary judgment was issued on Monday evening. One of the ways that courts decide whether copyrighted work is being used without permission in copyright law is by examining whether it was “transformative,” It is not the same as the original but it could be something else. “The technology at issue was among the most transformative many of us will see in our lifetimes,” Alsup wrote
“This is the first major ruling in a generative AI copyright case to address fair use in detail,” Chris Mammen is a partner in Womble Bond Dickinson and specializes in intellectual property. “Judge Alsup found that training an LLM is transformative use—even when there is significant memorization. He specifically rejected the argument that what humans do when reading and memorizing is different in kind from what computers do when training an LLM.”
In August 2024, the US District Court for Northern District of California received the first class-action lawsuit filed by authors alleging that Anthropic violated their rights by using their work without their permission.
Anthropic won the battle against artificial intelligence for the first time, but it comes with an asterisk. Alsup ruled Anthropic was using its training in a fair way, but he also ruled the authors can take Anthropic on trial for pirating.
Anthropic eventually switched to using purchased books for training, but it first built up and maintained a large library of pirated material. “Anthropic downloaded over seven million pirated copies of books, paid nothing, and kept these pirated copies in its library even after deciding it would not use them to train its AI (at all or ever again). Authors argue Anthropic should have paid for these pirated library copies. This order agrees,” Alsup is a writer.
“We will have a trial on the pirated copies used to create Anthropic’s central library and the resulting damages,” Order is complete.
Anthropic has not responded to our requests for comments. Attorneys representing the plaintiffs have declined to make any comments.
A lawsuit was filed. Bartz v. AnthropicAnthropic filed its first complaint on less than one year ago. In February, it requested summary judgement regarding the fair-use issue. Alsup is a judge with a lot more fair use experience than most federal judges, having presided over this first trial. Google v. OracleThe Supreme Court heard a case that was a milestone in the history of copyright, technology and tech.
