Security researchers from MATS Research, the ELLIS Institute Tübingen, the Max Planck Institute for Intelligent Systems, and Snyk have discovered that Anthropic, OpenAI, and Google use a single global encryption key across their entire AI ecosystem. By decoding 315,320 encrypted reasoning blocks from public GitHub and Hugging Face repositories, they recovered 182 credentials, including 62 live API keys, 33 passwords, and 30 personal email addresses. The vulnerability allows attackers to inject encrypted reasoning from a stronger model into a weaker sibling model to extract the plaintext reasoning. Anthropic, OpenAI, and Google deployed server-side patches after responsible disclosure, but historical session logs already shared publicly remain decodable.
Source: Read the original article
