China’s Alibaba will ban its employees from using Anthropic’s programming tool Claude Code starting July 10, according to multiple reports.
Anthropic has already banned the use of its models by Chinese companies and foreign entities owned by those companies. The company is reportedly working to close the loophole that allowed Chinese users to access Claude.
Part of closing that loophole included a version of Claude code that could secretly identify Chinese users, according to a recent Reddit post. Anthropic’s Thariq Shihipar said in a post on X that this was an “experiment we started in March with the goal of preventing account abuse by unauthorized resellers and protecting against distillation.” (Distillation is a practice in which AI models are trained based on the output of other models.)
“Since then, the team has put in place stronger mitigation measures. In fact, we’ve been meaning to do away with this for some time,” Schichper said.
Despite this, Alibaba is reportedly classifying Claude code as high-risk software and instructing its employees to use its Qoder tool instead.
