Anthropic has agreed to pay $1.5 billion to settle a lawsuit that accused the AI company of training its Claude model on pirated books. The settlement closes a legal fight that had drawn attention from regulators in Europe, who are now watching the case closely.
What the lawsuit alleged
The case claimed Anthropic used copyrighted books obtained from unauthorized sources to build Claude's training data. The plaintiffs argued that the company knowingly copied and processed the works without permission, violating copyright law. Anthropic did not admit wrongdoing as part of the settlement.
The settlement terms
The $1.5 billion payout is one of the largest ever in an AI copyright case. It covers a class of authors and publishers who said their books were used without consent. The money will be distributed among affected rights holders, though the exact allocation hasn't been disclosed.
European regulators take note
European authorities are monitoring the situation closely. The settlement could influence ongoing discussions in the EU about how AI companies should handle copyrighted material during training. Some policymakers have already called for stricter transparency rules on training data sources.
The case is a reminder that the race to build powerful AI models carries legal risks. Other companies facing similar lawsuits will be watching how this settlement shapes future negotiations and court rulings.
