The reported concern stems from claims made by a Reddit user who said they reverse-engineered Claude Code while restoring a disabled remote-control feature. In a technical write-up, the user alleged that versions of the tool beginning with version 2.1.91, released on April 2, checked whether a system’s proxy configuration or timezone matched entries on hidden internal lists.
According to that analysis, the lists included Chinese corporate networks, cloud regions, and AI organizations such as Alibaba, Baidu, ByteDance, and Moonshot AI. The report claimed that when a match occurred, Claude Code modified its own system prompt by changing the date format and replacing a punctuation character, rather than transmitting a separate telemetry signal.
Anthropic has not issued a formal public statement addressing those allegations. However, a member of the Claude Code team identified as Thariq reportedly said on social media that the mechanism was designed to limit account reselling and model distillation, adding that it would be removed in a future release. Reports from The Register and other outlets said work on that change was already underway by July 1.
If accurate, the reported mechanism remained in the software for roughly three months before being removed.
The reported ban also comes after escalating tensions between Anthropic and Alibaba over AI model security. In a June 10 letter to U.S. senators, Anthropic alleged that operators connected to Alibaba’s Qwen AI lab used nearly 25,000 fraudulent accounts to extract Claude’s software engineering and reasoning capabilities, generating more than 28.8 million exchanges between April 22 and June 5. Alibaba has not publicly responded to that accusation.
The latest report adds another chapter to that dispute, with allegations now extending beyond model distillation to the behavior of Anthropic’s own developer tooling. No independent security firm has published a full audit of the alleged detection mechanism, and it remains disputed whether it functioned as an anti-fraud safeguard or something more targeted.
Alibaba has not disclosed how the reported policy would be enforced internally. Reuters said its reporting was based on a single source, while neither Alibaba nor Anthropic provided an official statement for the report. As a result, the companies’ full positions on the reported July 10 policy remain unclear.
This analysis is based on reporting from TNW.
Image courtesy of Bloomberg.
This article was generated with AI assistance and reviewed for accuracy and quality.
