Alibaba Group has reportedly implemented a policy prohibiting its employees from using Anthropic's programming tool Claude Code, effective July 10, 2026.This decision follows Anthropic's existing restrictions on Chinese companies and entities controlled by them.

The company has been actively addressing loopholes that allowed Chinese users to access Claude, including a version that could secretly identify Chinese users.

Anthropic's Thariq Shihipar explained that an experiment launched in March aimed to prevent account abuse by unauthorized resellers and protect against distillation, a practice where AI models are trained on other models' outputs.The team has since implemented stronger mitigations and plans to remove the feature.Alibaba has classified Claude Code as high-risk software, directing employees to use its own Qoder tool instead.

The move underscores growing tensions between Chinese tech firms and Western AI developers, with security and data privacy concerns driving corporate policies.

This incident highlights the complex geopolitical landscape of AI technology, where companies navigate regulatory challenges and competitive dynamics.