Anthropic Settles Landmark Copyright Lawsuit with U.S. Authors
The terms of the settlement remain confidential.
AI startup Anthropic has reached a resolution in a high-stakes class action lawsuit brought by prominent U.S. authors, marking one of the most significant settlements in the ongoing battle over AI copyright infringement.
The lawsuit alleged Anthropic downloaded as many as 7 million pirated books to train its large language models, exposing the company to catastrophic damages.
Legal experts estimated potential penalties could have reached $900 billion if a jury ruled against the startup. By comparison, Anthropic is projected to generate just $5 billion in revenue this year while still operating at a loss.
In a court filing, Anthropic warned that the pursuit of billions in damages was a potential “death knell” for the company, forcing it to settle.
The terms of the settlement remain confidential, but the parties have been ordered to file for preliminary approval by September 5.
The agreement does not end Anthropic’s legal challenges. The company still faces lawsuits from Reddit over unauthorised scraping and Universal Music Group over alleged song lyric infringement—part of a broader wave of copyright disputes engulfing AI companies, including OpenAI, Meta, and Perplexity.
