A federal judge on Monday skewered a $1.5 billion settlement between artificial intelligence company Anthropic and authors who allege nearly half million books had been illegally pirated to train chatbots, raising the specter that the case could still end up going to trial.
After spending nearly an hour mostly lambasting a settlement that he believes is full of pitfalls, U.S. District Judge William Alsup scheduled another hearing in San Francisco on September 25 to review whether his concerns had been addressed.
“We’ll see if I can hold my nose and approve it” then, Alsup said before adjourning Monday’s hearing.
The judge’s misgivings emerged just a few days after Anthropic and attorneys who filed the class-action lawsuit announced a $1.5 billion settlement that is designed to resolve the pirating claims and avert a trial that had been scheduled to begin in December.
Alsup had dealt the case a mixed ruling in June, finding that training AI chatbots on copyrighted books wasn’t illegal but that Anthropic wrongfully acquired millions of books through pirate websites to help improve its Claude chatbot.
