Chinese tech giant Alibaba has banned its employees from using the Claude Code tool from Anthropic. The decision was made after the discovery of hidden code allegedly used to track users from China and employees of Chinese AI laboratories.
What Happened
As of July 10, Claude Code has been included in a list of programs with high risk levels and vulnerabilities. Alibaba restricted access to this tool within its corporate environments due to concerns regarding potential espionage and data compromise.
Context
The issue lies in the detection of hidden data collection mechanisms, which shifts the discussion of AI agent usage from efficiency to supply chain security. The incident highlights the risks of using proprietary tools that have access to sensitive codebases.
Why It Matters for the Industry
This incident intensifies the geopolitical fragmentation of the development tools market. An increased demand is expected for local, self-hosted, and open-source alternatives that can operate in isolated environments, as well as the formation of strict security audit standards for Agentic AI.
Why It Matters for Users
Developers and companies should exercise caution when using foreign AI agents in corporate environments. It is necessary to conduct thorough security audits for all tools that have access to source code and system calls.
What Is Not Yet Known / Limitations
At the moment, the focus of the discussion is shifting from purely engineering risks to macroeconomic and geopolitical consequences; however, the exact scale of the impact on global AI tool supply chains has yet to be assessed.
Sources
Author
Look at AI, Editorial Team