Alibaba has ordered employees to stop using Anthropic’s AI coding assistant, Claude Code, after raising concerns over alleged security risks linked to the software. The decision marks the latest flashpoint in the escalating AI battle between Chinese and American technology companies, where competition is increasingly spilling over into cybersecurity and intellectual property disputes.
Alibaba has reportedly banned employees from using Anthropic’s Claude Code in workplace environments, citing alleged security concerns surrounding features that could identify users connected to China.
According to Reuters, the internal ban will take effect on July 10 after the Chinese technology giant classified Claude Code as high-risk software. Employees have also been instructed to remove Claude-related software from their work computers and instead use Qoder, Alibaba’s in-house AI coding assistant.
The move comes amid growing tensions between Alibaba and Anthropic, whose relationship has deteriorated in recent weeks over allegations involving AI model theft, security concerns and restrictions on access to advanced AI systems. Alibaba’s decision followed scrutiny over code inside Claude Code that could help identify users operating from China or using China-linked networks. A person familiar with the matter said the company viewed the feature as presenting potential security risks for enterprise environments. Alibaba has not publicly confirmed the internal directive and did not respond to Reuters’ request for comment.
The controversy traces back to a reverse-engineering analysis published by a developer on Reddit, who claimed Claude Code contained mechanisms capable of checking system settings such as proxy configurations and time zones associated with Chinese organisations.
Anthropic has not described the feature as a backdoor. Instead, a member of the Claude Code engineering team said the mechanism was introduced to detect account reselling and large-scale AI model distillation rather than to spy on users. The company also indicated the feature would be removed in a future software update after concerns emerged.
The latest dispute arrives only weeks after Anthropic accused operators linked to Alibaba’s Qwen AI team of carrying out what it described as one of the largest known attempts to extract the capabilities of Claude through large-scale automated interactions.
According to Anthropic, the campaign involved approximately 25,000 fraudulent accounts generating more than 28 million interactions with Claude between April and June in an alleged effort to improve competing AI systems. Alibaba has not publicly responded to those allegations.
Claude Code has become one of Anthropic’s fastest-growing enterprise products. Unlike traditional AI chatbots, the command-line assistant helps software developers write, debug and manage code directly from their development environments, making it particularly popular among engineering teams. Although Anthropic officially restricts access to many of its services in mainland China, some developers have continued using Claude Code through overseas cloud servers and proxy services.
See Also: How Peter Thiel’s Top Lieutenant Is Quietly Finding the Next Billion-Dollar AI Startups
Alibaba’s latest directive effectively ends its use within the company’s workplace environment. The decision also highlights how AI competition is increasingly extending beyond model performance. Technology companies are now scrutinising the software, infrastructure and security mechanisms embedded inside AI products before allowing them onto corporate networks.
For Chinese firms, concerns about foreign-developed AI tools have grown alongside broader geopolitical tensions between Beijing and Washington.
At the same time, American AI companies have become increasingly cautious about protecting their models from unauthorised access and large-scale distillation attempts.
Industry analysts say those competing priorities are creating new challenges for global AI adoption.
Developers want access to the world’s most capable coding assistants, while enterprises must balance productivity with cybersecurity, compliance and intellectual property protection.
Alibaba’s decision reflects that growing tension. Rather than relying on Anthropic’s platform, the company is directing employees toward its internally developed AI coding assistant, reinforcing China’s broader strategy of reducing dependence on foreign AI technologies while accelerating investment in domestic alternatives.
The episode also illustrates how quickly trust can become a competitive advantage in the AI industry. As organisations deploy increasingly powerful coding assistants across sensitive software projects, confidence in how those systems handle user data may become just as important as the quality of the code they generate.
For Anthropic, the challenge now extends beyond building smarter AI models. Maintaining enterprise confidence in the security and transparency of its products may prove equally important as competition with Chinese AI companies continues to intensify.
About the Author
Marcel Chidozie is a tech analyst and writer covering foreign news, fintech, and emerging technologies at TechRegard. Based in Nigeria, He's passionate about translating complex tech developments into compelling, accessible stories for diverse audiences. His work focuses on how technology shapes innovation across Africa and globally.
