Anthropic Accuses Alibaba of Largest AI Distillation Attack to Date

Anthropic has accused Alibaba of conducting the largest known distillation attack on its Claude AI model. In a letter to US Senators, Anthropic detailed that operators linked to Alibaba executed nearly 29 million exchanges using approximately 25,000 fraudulent accounts between April 22 and June 5, 2026. This attack aimed to illicitly extract Claude's capabilities, including software engineering and reasoning skills. Anthropic described the operation as industrial-scale, warning that such tactics could help Chinese models reach advanced capabilities more quickly. The company urged Congress to implement stricter regulations and penalties against such activities. Alibaba has denied any wrongdoing and is currently involved in legal actions against the US government regarding its designation as a security risk. The incident has raised significant concerns about the implications for US technology and national security.

Key Points: • Anthropic claims Alibaba conducted the largest distillation attack on its Claude AI model. • The attack involved nearly 29 million exchanges through 25,000 fraudulent accounts. • Anthropic is calling for legislative action to prevent future distillation attacks.