Anthropic is accusing China's Alibaba of exploiting its AI models in a large-scale attack

Anthropic has accused Alibaba of conducting a large-scale “distillation attack” aimed at extracting capabilities from its Claude AI models. In a June 10 letter to U.S. senators Tim Scott and Elizabeth Warren, Anthropic’s head of policy, Sarah Heck, said Alibaba-affiliated operators tried to “illicitly extract Claude’s capabilities” to train models of their own. Heck alleged that between April 22 and June 5, the operators carried out 28.8 million exchanges with Claude through nearly 25,000 fraudulent accounts. She argued these actions were conducted systematically and at industrial scale to harvest U.S. AI capabilities without bearing the training and R&D costs of frontier models. The company requested stronger legislation, including limiting China’s access to advanced U.S. computing infrastructure and penalizing entities involved. The claim follows a U.S. export control on Anthropic’s Fable 5 and comes as Alibaba is also listed on a Pentagon blacklist; Alibaba later sued the U.S. government, and its shares were down more than 4% as of Thursday.