Chinese AI labs secretly used millions of Claude exchanges to train their models, Anthropic says

Imen Ben Youssef | Afp | Getty Images

Anthropic said it detected and disrupted unauthorized large-scale efforts by China-based AI labs including , Moonshot and DeepSeek to train their models using Claude.

The U.S. AI company said in a threat intelligence report

"Some of these exchanges included sensitive information, including from individual users, major multinational companies, and state-affiliated actors ... These practices are likely inconsistent with privacy laws and the labs' own terms of service," according to the report.

Anthropic said operators affiliated with Alibaba used Claude outputs to help train its Qwen models, while Moonshot routed some Kimi user requests to Claude and used some of the resulting exchanges to train its own models.

The Alibaba operation was the largest distillation campaign Anthropic said it has measured, involving more than 151 million exchanges with Claude between May and July.

The activity peaked at nearly 3 million exchanges per day from more than 3,500 fraudulent accounts, according to the report. The company said Alibaba also used Claude for broader AI research, including reinforcement learning and model architecture.

Moonshot and DeepSeek

Anthropic also detailed activity involving Moonshot AI, the Beijing-based company behind the Kimi family of AI models.

Moonshot silently forwarded some customer requests intended for Kimi to Claude and then displayed Claude's responses to users, who thought they were using a Kimi model, according to the report.

In one 10-day period, Moonshot relayed nearly 300,000 customer requests to Anthropic, the vast majority of which were routed to Claude Opus models. The requests were routed through a network of 5,380 accounts that Anthropic described as fraudulent, most of which appeared to be located in Singapore and Japan.

The report said Moonshot saved at least some of those exchanges and extracted Claude's reasoning transcripts to use as training data for its own models.

More than 23 million exchanges were attributed to Moonshot between May and July, according to the report.

Some of the customer requests routed to Claude contained sensitive information. The company said it did not know whether Moonshot had notified customers that their requests were being sent to Anthropic.

The company said DeepSeek — which rose into prominence last year due to its capabilities and cheap costs — also used tactics similar to Moonshot, transferring exchanges to Claude without notifying DeepSeek customers. Anthropic said it observed more than 12 million distillation attacks attributable to DeepSeek over 14 days in July 2026.

The report, which named several other major Chinese AI companies, covers activity the company said it disrupted between December 2025 and August 2026 across seven areas, including cyber operations, influence operations, surveillance, scams and fraud, biological misuse, conventional weapons development and distillation.

Alibaba, Moonshot, DeepSeek, Xiaomi and Anthropic did not immediately respond to CNBC's requests for comment.