Anthropic sent a letter to the U.S. Senate Banking Committee and the White House, accusing Alibaba's Qwen Lab of launching "the largest known distillation attack to date" against Claude.

According to the letter obtained by Bloomberg, between April 22 and June 5, operators associated with Alibaba interacted with Claude over 28.8 million times through nearly 25,000 fake accounts, targeting Claude's most valuable capabilities, including software engineering and agent reasoning.

Anthropic stated that these attacks were carried out "illegally, systematically, and on an industrial scale, with the goal of harvesting U.S. AI capabilities and repackaging them as their own products, without bearing the R&D costs required to train frontier models."

Anthropic, OpenAI, and Google have jointly established an information-sharing mechanism to track distillation activities that violate terms of service. Senators Bill Hagerty and Andy Kim plan to introduce an amendment as soon as this week, adding Chinese companies that illegally distill U.S. AI models to sanctions or blacklists, attached to the must-pass National Defense Authorization Act.