Alibaba Bans Claude Code Over Hidden Tracking Code Discovered in Anthropic's Developer Tool
Key Takeaways
- ▸Alibaba has instituted a company-wide ban on Claude Code effective July 10, 2026, citing security vulnerabilities and back-door risks
- ▸Anthropic embedded hidden code in Claude Code designed to track users based in China or affiliated with Chinese AI research labs
- ▸Security researchers discovered and publicly disclosed the hidden tracking code on Reddit and GitHub, triggering widespread backlash
Summary
Alibaba Group has banned employees from using Anthropic's Claude Code developer tool, citing serious security vulnerabilities. The ban, effective July 10, 2026, follows the discovery of hidden code embedded in Claude Code that could track whether users were based in China or affiliated with Chinese AI research organizations. Alibaba classified the tool as 'high-risk software' in an internal notice to staff.
Security researchers discovered the hidden tracking functionality and published their findings on platforms including Reddit and GitHub earlier this week. The covert code represented a significant breach of user trust, as it operated without users' knowledge or consent. The discovery has sparked widespread backlash across the tech community and prompted questions about the integrity of AI development tools and supply chain security.
For Anthropic, the ban represents a major business setback in China, one of the world's largest and most strategic AI markets. The incident raises broader concerns about embedded surveillance in developer platforms and potential espionage risks in cross-border AI competition and development.
- The ban highlights critical security concerns around AI developer tools and escalating tensions in cross-border AI competition
