Big tech’s security reviews of AI tools are getting increasingly strict, and Alibaba’s ban is likely to prompt many companies to follow suit.

Starting July 10, Alibaba has banned employees from using Anthropic's Claude in work environments, citing security risks of an embedded backdoor, applicable to all internal environments. Although no official statement has been released, sources say the decision was based on a security risk assessment and reflects financial institutions' concerns about the integration of AI tools in development, research, and financial services processes, as well as Claude's support status in Hong Kong and related bank restrictions.