Anthropic says Alibaba illicitly extracted Claude AI model capabilities
Details of the Alleged Extraction and Industry Response
By Karen Freifeld
June 24 (Reuters) - U.S. AI company Anthropic accused Alibaba, the Chinese technology and e-commerce giant, of illicitly extracting its Claude AI model capabilities in what it said was the largest known attack of its kind on the company, according to a letter seen by Reuters.
Nature of the Attack
The strike by Alibaba is described as a "distillation" effort, which Anthropic has said involves training a less capable model on the outputs of a stronger one.
Scope and Timeline of the Campaign
Anthropic said the campaign was conducted between April 22 and June 5, 2026, and generated more than 28.8 million exchanges with Claude through almost 25,000 fraudulent accounts.
Purpose and Affiliations
Anthropic said in the letter that distillation is a way to help accelerate China's ability to reach Anthropic's advanced Mythos Preview capabilities.
It said the campaign was conducted by operators affiliated with Alibaba and Alibaba Qwen, Alibaba's AI lab. Alibaba did not immediately respond to a request for comment.
Official Correspondence and Government Involvement
The letter, dated June 10, was sent to Senators Tim Scott and Elizabeth Warren, the chair and ranking member, respectively, of the U.S. Senate Banking Committee, ahead of a scheduled hearing on AI.
Context of U.S.-China AI Tensions
In April, the White House accused China of stealing U.S. AI labs' intellectual property on an industrial scale.
Anthropic's Support for U.S. Government Efforts
Anthropic said in the letter it was supportive of the U.S. government's efforts to combat the attacks, including partnering with private sector AI companies through threat-intelligence sharing and other exercises.
Other Incidents Involving Chinese AI Labs
Anthropic said in a February posting that it had identified a campaign by Chinese AI startup DeepSeek — whose low-cost AI model sent shockwaves through the technology world in January 2025 — and two other Chinese AI labs to illicitly extract capabilities from its Claude AI platform.
Scale of Operations by DeepSeek and Others
It said DeepSeek's operation involved over 150,000 exchanges, while Moonshot AI was at a scale of over 3.4 million and MiniMax over 13 million.
Growing Threat and Industry Response
It also said at the time that the campaigns were growing in "intensity and sophistication" and that addressing the threat would require "rapid, coordinated action among industry players, policymakers and the global AI community."
U.S. Government Actions and Ongoing Concerns
Alibaba's Military Designation
Alibaba was added to the Pentagon's Chinese military companies list this month, a designation it is challenging.
Commerce Department Decisions
But the Commerce Department has held off placing DeepSeek on a trade blacklist, as Reuters exclusively reported this month, despite it being deemed a national security risk by an interagency governmental committee, as the department tries to avoid escalating tensions with Beijing.
Restrictions on Anthropic's AI Models
Meanwhile, on June 12, two days after Anthropic sent the letter, the Commerce Department imposed controversial restrictions on Anthropic's latest Mythos and Fable AI models because officials feared they could be deployed by military intelligence users in China and other countries of concern. The restrictions resulted in Anthropic disabling access to the models globally.
Reporting Credits
(Reporting by Karen Freifeld; additional reporting by David Shepardson; Editing by Sonali Paul)
