Alibaba bans Claude Code after Anthropic backdoor concerns
Synopsis
Alibaba has banned Claude Code across its workforce from July 10, after security researchers exposed hidden tracking code in Anthropic's AI coding tool that allegedly identified users based in China — a rare public corporate response that could reshape how Chinese firms evaluate US AI software.
Key Takeaways
Alibaba Group Holding issued an internal notice on Thursday, July 3, 2026 banning all staff from using Anthropic 's Claude Code effective July 10 .
Alibaba classified Claude Code as 'high-risk software with security vulnerabilities' after it was reportedly found to contain code capable of identifying users based in China or affiliated with Chinese AI laboratories.
Security researchers discovered the alleged backdoor earlier this week and published findings on Reddit and GitHub , triggering widespread backlash.
Anthropic did not immediately respond to requests for comment on the allegations.
The ban is among the most prominent corporate responses to the controversy and signals growing scrutiny of foreign AI developer tools inside China 's technology sector.
Alibaba Group Holding has prohibited its employees from using Anthropic's Claude Code coding agent, citing security vulnerabilities linked to the US AI firm's alleged use of hidden tracking code targeting Chinese users — a disclosure that has triggered significant backlash across the tech community since earlier this week.
The internal ban
In an internal notice dated Thursday, July 3, 2026, Alibaba described Claude Code as carrying 'back-door risks' and confirmed it had been added to a list of high-risk software with security vulnerabilities. 'As Claude Code was recently discovered to carry back-door risks, after comprehensive evaluation, Claude Code has now been added to a list of high-risk software with security vulnerabilities,' the notice read, according to reports. All staff members will be prohibited from using Claude Code in the office starting from July 10.
What Anthropic allegedly did
The ban stems from Anthropic's earlier move to embed code within its Claude Code platform that could secretly determine whether a user was based in China or affiliated with a Chinese AI laboratory, according to people familiar with the matter. Security researchers uncovered the covert practice earlier this week, publishing their findings on platforms including Reddit and GitHub. The revelations set off widespread criticism of Anthropic's practices among developers and security professionals.
Why it matters
The episode highlights the deepening distrust between US and Chinese technology ecosystems at a time when AI coding tools have become integral to enterprise software development. For a company of Alibaba's scale — one of China's largest technology employers — a blanket prohibition on a widely used AI developer tool signals that geopolitical and cybersecurity concerns are now directly shaping workplace software policy. Other Chinese technology firms are likely to review their own exposure to Anthropic's products in the wake of this incident.
The competitive backdrop
Anthropic's Claude Code had gained traction among developers globally as a capable AI-assisted coding platform, competing with tools from OpenAI and others. The alleged tracking behaviour, if confirmed, would represent a significant breach of user trust and could accelerate China's push toward domestically developed AI developer tools. Industry analysts note that incidents like this tend to reinforce the bifurcation of global AI infrastructure along geopolitical lines.
What's next
Anthropic did not immediately respond to requests for comment. Alibaba similarly did not issue a public statement beyond the internal notice. As scrutiny intensifies, the incident is expected to prompt broader regulatory and enterprise-level reviews of foreign AI tools used in Chinese workplaces — and may accelerate calls in Washington and Beijing alike for clearer rules governing AI software deployed across borders.
Point of View
Functionally, a form of competitive intelligence gathering, regardless of intent. For Chinese enterprises, the incident validates long-standing arguments for AI sovereignty and will likely fast-track procurement policies favouring domestic alternatives. The episode also puts Anthropic in an uncomfortable position globally — developer trust, once eroded, is difficult to rebuild, and the company's expansion ambitions in non-US markets now face a credibility headwind that no product update can easily reverse.
NationPress
3 Jul 2026
Frequently Asked Questions
Why did Alibaba ban Claude Code?
Alibaba banned Claude Code because security researchers discovered that Anthropic 's AI coding tool allegedly contained hidden code capable of identifying whether users were based in China or linked to Chinese AI laboratories. Alibaba classified the tool as high-risk software with security vulnerabilities in an internal notice dated July 3, 2026 .
What is the Anthropic Claude Code backdoor controversy?
Security researchers found that Anthropic had embedded code in its Claude Code platform that could covertly determine if a user was located in China or affiliated with a Chinese AI lab. The findings were published on Reddit and GitHub earlier this week, prompting widespread criticism from developers and security professionals.
When does Alibaba's Claude Code ban take effect?
Alibaba 's ban on Claude Code takes effect on July 10, 2026 . The company issued the directive through an internal notice on Thursday, July 3, 2026 , instructing all employees to stop using the tool in the office by that date.
How does this affect Anthropic's business in China?
The ban signals a significant trust deficit for Anthropic among Chinese technology enterprises and could prompt other Chinese firms to audit or restrict Anthropic products. It is likely to accelerate demand for domestically developed AI coding tools and may complicate Anthropic 's broader international growth ambitions.
Who discovered the Anthropic tracking code in Claude Code?
Independent security researchers discovered the alleged tracking code embedded in Anthropic 's Claude Code platform earlier this week. They shared their findings publicly on platforms including Reddit and GitHub , where the revelations quickly gained attention and sparked backlash within the developer community.
