San Francisco-based Anthropic has accused Chinese e-commerce and tech giant Alibaba of extracting the capabilities of its Claude AI model through a large-scale distillation attack.
In a letter to US Senators Tim Scott and Elizabeth Warren, the company revealed that Alibaba-linked operators created thousands of fraudulent accounts to carry out nearly 29 million exchanges with Claude. The goal was to steal advanced functions like complex reasoning and decision-making.
Anthropic called on Congress to penalize the perpetrators and strengthen measures to prevent the theft of American technology. The company warned that such attacks turn hundreds of billions of dollars in US research and development into a subsidy for geopolitical rivals.
Alibaba has not responded to the allegations.