Anthropic has accused Qwen of using 25,000 fake accounts to extract nearly 29 million transactions with Claude, the largest such campaign to date. Anthropic has accused Alibaba of conducting the largest distillation campaign to date against a US artificial intelligence company, informing senators and White House officials that operators linked to Qwen AI used nearly 25,000 fake accounts to extract Claude's capabilities between April and June.
See also: Microsoft accuses Anthropic of presenting Claude as conscious
The letter, a copy of which was seen by Bloomberg, described nearly 29 million exchanges with Claude that targeted software engineering and practical logic, the model’s most commercially valuable skills. The accusation marks the first time Anthropic has named a major Chinese tech giant as the source of a distillation attack.
Previous charges in February targeted smaller Chinese AI startups, including DeepSeek, MiniMax and Moonshot AI, which Anthropic said had collectively generated more than 16 million exchanges through about 24,000 fake accounts. Alibaba’s campaign alone surpassed the combined volume of all three previous efforts. Distillation is the practice of feeding carefully crafted queries to a leading AI model, collecting its responses and using those responses to train a cheaper competing system that approaches the original’s capabilities.
The White House labeled the technique a national security issue in April, when OSTP Director Michael Kratsios released a memo committing the government to sharing information with U.S. AI labs about foreign disinformation campaigns. Anthropic said in its letter that Alibaba’s campaign was carried out after the Kratsios memo, defying government warnings.
See also: Anthropic expands Claude's security reach with 28 new integrations
Alibaba had no comment on the allegations. An Anthropic spokesperson declined to discuss specifics but stressed the importance of combating distillation through coordinated action between government and industry.
Alibaba’s American depositary receipts fell more than three percent on the news, falling below $100 in afternoon trading on Wednesday. The stock’s decline adds to a difficult period for the company in Washington, where it faces pressure on multiple fronts. The Pentagon added Alibaba to its list of Chinese military companies on June 8, a designation Anthropic cited in its letter.
Alibaba sued the Department of Defense this week to win removal from that list, calling the label baseless and claiming it has no military ties. The distilling lawsuit now opens a second front, framing Alibaba not only as a company with alleged military ties but also as an active participant in what Anthropic calls the systematic theft of American AI capabilities.
In its letter, Anthropic warned that hostile distillation allows Chinese labs to replicate cutting-edge AI at a fraction of the cost of training, and that the models built this way often lack safeguards. The company called on the Trump administration to clarify antitrust guidelines so that American labs can share more information about distillation efforts, reiterated its support for export controls on advanced AI chips, and called for sanctions against companies that use the technique.
See also: Claude Fable 5: Anthropic releases first public Mythos-Class model
Lawmakers are moving in tandem. Senators Bill Hagerty and Andy Kim plan to introduce an amendment to the defense bill that must pass that would blacklist or sanction any Chinese company found to have improper access to the output of American AI models. A related bipartisan bill in the House, sponsored by Representatives Bill Huizenga and Sydney Kamlager-Dove, is also being considered, although whether any of the proposals will survive the final version of the defense bill is uncertain.
