Alibaba bans Anthropic Claude AI for security concerns

Alibaba Group will ban its employees from using all artificial intelligence products from the U.S. company Anthropic, including the Claude Code assistant, starting July 10, according to sources familiar with the decision. The Chinese technology giant cited security concerns, specifically the risk of embedded backdoors that could allow unauthorized access to its systems.

The directive, which has not been officially confirmed by Alibaba, marks a sharp reversal from a previous company policy. Since early 2026, Alibaba had encouraged engineers to use third-party AI tools, even offering a reimbursement program that allowed developers to claim expenses for using services like Claude Code. That program will now end, pushing employees toward domestic alternatives, including Alibaba’s own Qwen model family.

The ban follows a period of escalating tensions between the two companies. In a letter to U.S. senators dated June 10, Anthropic accused operators linked to Alibaba’s Qwen AI lab of conducting what it called the largest data extraction campaign ever discovered against its models. Anthropic alleged that approximately 25,000 fraudulent accounts were used to generate over 28.8 million...