Alibaba will prohibit employees from using Anthropic’s artificial intelligence tools for work purposes from 10 July, citing security concerns, according to a CNBC report.
The Chinese technology company has reportedly classified Anthropic’s Claude Code as high-risk software and instructed employees to remove Anthropic’s AI models and agent products from their work systems. Employees have instead been directed to use Alibaba’s in-house AI assistant, Qoder.
The move comes weeks after Anthropic accused Alibaba of attempting to improperly extract its AI capabilities. In a letter submitted to the US Senate Committee on Banking, Housing, and Urban Affairs in June, Anthropic alleged that Alibaba had carried out what it described as the largest known “distillation attack” against the company. Distillation refers to techniques used to replicate or learn from the outputs of another AI model.
Anthropic’s terms of service prohibit companies from China and certain other countries from accessing its AI models. The company has also been tightening measures to prevent organisations from bypassing these restrictions.
According to recent reports, Anthropic has been working to close loopholes that allegedly allowed some Chinese companies to access Claude through overseas entities or third-country networks.
The latest decision also follows growing discussion on online developer forums, where users claimed Anthropic had embedded mechanisms to detect whether users were accessing its services from China.
Separately, reports have suggested that some Chinese technology companies have adopted alternative approaches to enable employees to learn from global AI tools. While TikTok parent ByteDance reportedly does not provide direct access to Claude, it has introduced a reimbursement programme allowing engineers to claim expenses for personal subscriptions used for skill development.
Alibaba and Anthropic have not publicly commented on the latest development.
