Anthropic has secured court approval for a 1.5 billion dollar settlement over its use of pirated books to train its Claude AI model. The settlement covers over 482,000 works, with approximately 91% already claimed by their authors or publishers, averaging around 3,000 dollars per book. This amount represents the first settlement of this magnitude among the dozens of copyright lawsuits related to generative AI, a precedent that directly affects OpenAI, Meta and Stability AI. The company resolved this case while preparing for its IPO with a private valuation near 965 billion dollars, thereby avoiding displaying this legal risk in its public prospectus.
Source: Read the original article
