Alibaba has officially moved to restrict its workforce from utilizing Anthropic’s AI-powered programming tool, Claude Code. The directive, which is set to take effect on July 10, classifies the software as a high-risk application. Employees are being directed to transition their development workflows to the company’s proprietary alternative, known as Qoder.

This decision follows ongoing friction regarding the accessibility of Anthropic’s models within China. Anthropic maintains strict policies that prohibit Chinese companies and their foreign subsidiaries from accessing its AI services. The company has been actively working to identify and close technical loopholes that have previously allowed users in restricted regions to bypass these geographic and corporate barriers.

Recent reports suggest that Anthropic previously implemented experimental measures to detect unauthorized access, including a version of Claude Code designed to identify users based in China. While Anthropic representatives have characterized these efforts as temporary experiments aimed at preventing account abuse and protecting against model distillation, the company has since moved toward more robust, permanent mitigation strategies. In response to these tightening restrictions, Alibaba is prioritizing the adoption of its internal AI tools to ensure compliance and maintain control over its development environment.

Key Takeaways

• Alibaba has officially designated Anthropic’s Claude Code as high-risk software, banning its use among employees.
• The restriction aligns with Anthropic’s existing policy of prohibiting Chinese entities from accessing its AI models.
• Alibaba is mandating that staff transition to its own internal programming tool, Qoder, to replace the restricted software.

Editor’s Analysis & Impact

The move by Alibaba to ban Claude Code highlights the deepening fragmentation of the global AI landscape, driven by both geopolitical tensions and corporate security concerns. As major AI developers like Anthropic tighten access to their models, large Chinese technology firms are being forced to accelerate the development of sovereign AI stacks. This shift is likely to result in a bifurcated ecosystem where Western and Chinese AI tools operate in distinct silos. For the industry, this underscores the growing importance of ‘AI sovereignty’—the ability for a company or nation to rely on its own proprietary models rather than depending on foreign-developed infrastructure that may be subject to sudden access restrictions or regulatory changes. Moving forward, we expect to see increased investment in domestic AI alternatives as companies seek to insulate their operations from international policy shifts.

Frequently Asked Questions

Q: Why is Alibaba banning Claude Code?
A: Alibaba has classified Claude Code as high-risk software, likely due to Anthropic’s strict policies against Chinese companies using its models and the potential for regulatory or security complications.

Q: What tool will Alibaba employees use instead of Claude Code?
A: Alibaba is instructing its employees to transition to the company's own proprietary programming tool, Qoder.