Anthropic swiftly removed a hidden tracker from its Claude Code tool after a security researcher revealed it was secretly monitoring users in China. The tracker, which used a technique called prompt steganography to conceal its code, collected data on users' timezone, proxy, and potential connections to Chinese AI labs that Anthropic has accused of distillation attacks.
An Anthropic engineer stated the tracker was added as an experiment in March to prevent account abuse and unauthorized reselling. Reports show that unauthorized resellers have been selling access to Anthropic’s models at significantly reduced prices.
Comments