China warns of “security backdoor” in Anthropic AI coding tool

China’s National Vulnerability Database (NVDB) has warned that versions of Anthropic’s AI coding tool, Claude Code, may contain a security backdoor that could transmit sensitive information, including location and identity data, back to Anthropic’s servers without user consent.
The NVDB, affiliated with China’s Ministry of Industry and Information Technology, reported that the AI tool poses a severe threat and advised immediate checks and updates to the latest secure versions to mitigate the risks.
Anthropic has not yet responded to requests for comment on these allegations, which first surfaced in tech media.
To address these concerns, Chinese tech giant Alibaba has banned the use of Claude Code effective July 10, citing security issues.
An Anthropic engineer has stated that the potentially problematic data tracking was an experimental measure to deter account abuse by unauthorized resellers and mitigate reverse-engineering efforts, like those previously accused of by Anthropic towards Alibaba.
The engineer assured that stronger mitigations have been implemented and the measure in question should be fully revoked in the next release.

You may have missed