Alibaba bans Claude AI after secret user tracking scandal

Alibaba banned Anthropic's Claude AI after discovering undisclosed tracking of Chinese users, citing security and industrial espionage concerns.

Alibaba has implemented a complete ban on Anthropic's Claude AI tools for its employees after uncovering undisclosed tracking mechanisms in the Claude Code product. Security researchers revealed that Anthropic embedded steganographic markers to monitor users suspected of connecting from China, collecting data such as time zones and proxy settings without user consent. These tracking features were only removed after public exposure. Anthropic defended its actions by citing large-scale attempts, allegedly linked to Alibaba, to extract Claude's capabilities and improve rival AI models. Reports indicate nearly 25,000 suspicious accounts and millions of unauthorized exchanges. The incident has heightened US-China tech tensions, raising concerns about industrial espionage, software security, and user trust. Alibaba has now classified Claude Code as high-risk, requiring employees to uninstall it and switch to its own AI assistant, Qoder. Both companies have declined to comment publicly.