Alibaba Bans Employees From Using Claude Code Amid Tech Tensions

Alibaba has officially categorized Anthropic’s Claude Code as high-risk software, ordering its workforce to cease usage by July 10. The company is steering staff toward its internal coding solution, Qoder, to mitigate potential data security concerns. This move follows a tightening of access restrictions between the Chinese tech giant and the US-based AI startup, which has previously attempted to block Chinese entities from utilizing its advanced models.

The decision reflects a broader intensification of the US-China artificial intelligence race. While Western firms move to limit access to sensitive programming tools, Chinese firms are rapidly deploying high-performance local alternatives like the GLM-5.2 model. This Beijing-based innovation has drawn international attention for its agent-based task execution capabilities, with industry analysts comparing its emergence to the disruptive rise of previous domestic AI breakthroughs.

Anthropic recently reported that the US Department of Commerce lifted export restrictions on its flagship Claude Fable 5 and Mythos 5 models. Despite these regulatory changes, the internal crackdown within Alibaba indicates that major Chinese firms remain cautious about relying on foreign AI infrastructure. Future developments will hinge on how effectively Chinese developers bridge the performance gap between domestic tools and Western frontier models, directly impacting the operational workflows of the country’s massive tech labor force.