Anthropic accuses Alibaba of large‑scale distillation attack to copy Claude AI model

US‑based AI firm Anthropic has filed a complaint alleging that Chinese technology giant Alibaba conducted an extensive “distillation” operation against its Claude models. According to Anthropic, Alibaba used roughly 25,000 fake accounts to generate more than 28.8 million prompts between 22 April and 5 June 2026, aiming to capture the advanced reasoning and other capabilities of Claude, including the Mythos Preview model. The company says the activity not only breaches policy but also poses a national‑security threat to the United States.

Following Anthropic’s letter, the U.S. Department of Commerce imposed export controls on Anthropic’s Mythos and Fable models, citing concerns over their use by foreign entities. Anthropic has previously raised similar accusations against other Chinese AI firms such as DeepSeek, Minimax and Moonshot AI.