AI distillation crackdown: Anthropic is urging Congress to act against Chinese rivals it says engaged in large-scale model distillation from its Claude chatbot, calling for policy responses at the federal level. Anthropic alleges more than 28.8 million exchanges with Claude occurred between April 22 and June 5 and that nearly 25,000 fraudulent accounts were used in the campaign. Anthropic set out these allegations in a June 10 letter to leaders of the Senate Banking, Housing, and Urban Affairs Committee.
Anthropic alleges that Alibaba-affiliated operators generated more than 28.8 million exchanges with its Claude chatbot between April 22 and June 5, using nearly 25,000 fraudulent accounts. The operations targeted Claude’s agentic reasoning, software engineering, and long-horizon planning capabilities. Anthropic characterizes the activity as a large-scale distillation campaign conducted over a multiweek period. The company set out these allegations in a June 10 letter to Senate Banking, Housing, and Urban Affairs Committee leaders.
Anthropic warned that the distillation campaign could accelerate China’s military and cyber AI capabilities and narrow the United States’ technological lead. “Beyond its scale, this campaign was striking for its brazen nature,” Anthropic wrote. “When PRC labs distill these capabilities from U.S. models, they capture the returns on American investments without bearing the costs or risks associated with training frontier AI models,” Anthropic wrote.
Anthropic described the distillation campaign as a national security concern that could accelerate China’s military and cyber AI capabilities and narrow the United States’ technological lead. The company urged expanding intelligence sharing and clarifying antitrust rules for sharing distillation-attack information as part of a coordinated government and industry response. Anthropic also called for strengthening export controls on advanced AI chips and compute, closing overseas data-center access loopholes for Chinese firms, and imposing penalties for large-scale model extraction.
Anthropic presented these recommendations in a June 10 letter to Senate Banking, Housing, and Urban Affairs Committee leaders and framed the activity as a large-scale, multiweek campaign. The company warned that distillation by foreign labs captures the returns on American investments without bearing training costs and risks. “Beyond its scale, this campaign was striking for its brazen nature,” Anthropic wrote. The company’s filing came after President Donald Trump signed an executive order earlier in the month expanding AI-powered cybersecurity initiatives.
Anthropic asked Congress and the administration to consider the specific measures it outlined to address the risks it identified. The company communicated these concerns directly to Senate committee leadership.
Anthropic sent a letter dated June 10 to Senate Banking, Housing, and Urban Affairs Committee Chairman Tim Scott and Ranking Member Elizabeth Warren that outlined its allegations about a large-scale distillation campaign targeting its Claude chatbot and requested policy responses. The company presented its concerns and recommendations directly to congressional leaders and framed the outreach as part of ongoing engagement with lawmakers and officials.
“We believe combating the threat of illicit distillation requires coordinated action between government and industry, and we will continue working with Congress and the administration to maintain American AI leadership.” A spokesperson for Anthropic declined to comment specifically on the June 10 letter.
Anthropic has urged an AI distillation crackdown after identifying what it says was a large-scale campaign that used fraudulent accounts to extract capabilities from its Claude chatbot, characterizing the activity as systematic and extensive. The company framed the operation as raising national security concerns and urged congressional and executive action to address those risks through coordinated policy measures.
