Anthropic is calling on Congress to strengthen protections against AI model distillation after claiming that Alibaba-affiliated operators carried out the largest known effort to extract capabilities from Claude.
In a June 10 letter to Senate Banking Committee Chairman Tim Scott and Ranking Member Elizabeth Warren, Anthropic alleged that operators tied to Alibaba and its Qwen AI lab generated more than 28.8 million exchanges with Claude between April 22 and June 5. They used nearly 25,000 fraudulent accounts — not real users.
Distillation attacks involve extracting a model’s capabilities to reproduce its behavior without the cost of training a frontier system. Anthropic said this campaign targeted Claude’s agentic reasoning, software engineering, and long-horizon planning abilities.
“Beyond its scale, this campaign was striking for its brazen nature,” Anthropic wrote. The company noted that Alibaba is listed on the NYSE, maintains US operations, and is accountable to US investors and regulators.
Anthropic framed large-scale distillation as a national security issue and urged lawmakers to expand intelligence sharing, strengthen export controls, and penalize companies responsible for unauthorized model extraction.
