Alibaba to ban employees from using Anthropic's coding tool: Report
The move follows allegations that the coding tool hid a backdoor tied to Chinese networks and cloud regions, Reuters reported.
- Alibaba will ban employees from using Anthropic's Claude Code in company workspaces starting July 10 due to allegations of hidden detection targeting Chinese users.
- A Reddit user claimed to have found code obfuscation related to user detection in Claude Code during the restoration of remote-control features.
- Version 2.1.91 of Claude Code checked for Chinese timezones or proxies linked to Chinese tech firms such as Alibaba, Baidu, and ByteDance, modifying system prompts subtly.
- Anthropic stated the mechanism was to prevent account selling and model distillation, promised to remove it, and accused Alibaba's Qwen AI lab of using 25,000 fake accounts to access Claude's capabilities.
25 Articles
25 Articles
The Chinese technology giant Alibaba banned its employees from using the Claude Code, from Anthropic, at work, after the tool is analyzed by resources that can help identify users linked to China, according to a person familiar with the determination. The ban is part of a growing dispute between the two companies, after Anthropic accused Alibaba of illegally extracting the capabilities of its model of IA Claude — a dispute that highlights the fr…
Alibaba Bans Employees From Using Claude
Alibaba Group has banned employees from using Anthropic’s Claude Code, and asked them to remove all Claude models from their work computers, according to two employees with knowledge of the matter. Alibaba told some employees Friday that the ban is due to security concerns about Anthropic, after a recent controversy over Claude Code’s hidden code that was capable of tracking whether the user
Coverage Details
Bias Distribution
- 61% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
