Anthropic alleges Alibaba ran AI distillation attack

The U.S. artificial intelligence company Anthropic has accused Chinese technology giant Alibaba of orchestrating a massive, illicit campaign to extract capabilities from its Claude AI model. In a letter sent to U.S. senators and White House officials, Anthropic stated that operators linked to Alibaba's Qwen AI lab used nearly 25,000 fraudulent accounts to conduct approximately 28.8 million exchanges with Claude between April and June of this year.

Anthropic described the operation as the largest known "distillation attack" on its systems to date. This technique involves systematically querying a sophisticated AI model to generate training data for a competing, less advanced system, allowing the competitor to replicate capabilities at a fraction of the development cost. The company said the campaign specifically targeted Claude's most commercially valuable functions, including software engineering and agentic reasoning.

The letter, dated June 10 and addressed to Senators Tim Scott and Elizabeth Warren among others, alleges the activity continued despite a White House memo in April directing agencies to crack down on large-scale exploitation of U.S. AI models through proxy...