Alibaba bans employees from using Anthropic's Claude Code

Alibaba will ban employees from using Anthropic’s programming tool Claude Code starting on July 10. The company has classified Claude Code as high-risk and is instructing staff to use its own Qoder tool instead.

What exactly is being banned and when?

Alibaba is prohibiting employees from using Claude Code beginning July 10, and the company has labeled the Anthropic tool as high-risk while directing workers to use Alibaba’s internal Qoder tool. The move is a company-level instruction to replace Claude Code with an internal alternative rather than to suspend all AI tool usage.

The ban follows multiple reports and internal guidance; the public details provided so far identify the effective date, the classification of Claude Code as high-risk, and the recommended replacement, Qoder.

Why did Alibaba classify Claude Code as high-risk?

Alibaba’s action follows concerns over access controls and a reported version of Claude Code that could identify Chinese users, and it comes amid broader limits Anthropic places on Chinese companies. Anthropic already prohibits Chinese companies, and foreign entities owned by those companies, from using its models, and the company has been working to close loopholes that let Chinese users access Claude.

A recent Reddit post claimed some of those mitigation efforts involved a version of Claude Code that could secretly identify Chinese users. Anthropic’s Thariq Shihipar addressed the claim on X, calling it “an experiment we launched in March that was meant to prevent account abuse from unauthorized resellers and protect against distillation.” Shihipar added that “the team has landed stronger mitigations since then and we’ve actually been meaning to take this down for a while.”

Those statements frame the technical and policy context: Anthropic has restrictions on who may use its models, it ran an experiment in March to curb misuse, and Alibaba appears to have judged the result — or the prior access issues — serious enough to restrict employee access.

Why it matters

The ban highlights how companies are reacting to both model-provider rules and perceived security risks inside large organizations. Alibaba’s decision affects developer workflows by shifting employees from an externally provided tool to an in-house option, Qoder, and it underscores the friction that arises when a cloud of policy, enforcement and technical mitigation surrounds model access.

It also sheds light on how vendors and customers manage jurisdictional restrictions: Anthropic’s policy against Chinese companies using its models, plus its efforts to close access loopholes, intersect directly with enterprise procurement and internal security policies at major Chinese firms.

What to watch

Watch whether Anthropic follows through on removing the March experiment and how Alibaba enforces the July 10 deadline. The two concrete signals to track are whether the reported mitigation that could identify Chinese users is taken down, as Thariq Shihipar said the team has been "meaning to take this down for a while," and whether Alibaba’s Qoder rollout replaces Claude Code usage in practice across teams.

If Alibaba publishes any formal internal guidance or if Anthropic updates its access controls, those will be the next public clues about how this dispute over access, mitigation and corporate policy will settle.