Google catches first AI-generated zero-day exploit in the wild
Google's Threat Intelligence Group disclosed the first known AI-built zero-day exploit caught in the wild. A cybercrime group used a language model to write a Python script that bypassed two-factor authentication in a popular open-source server administration tool. The vendor pat
Google's Threat Intelligence Group disclosed the first known AI-built zero-day exploit caught in the wild. A cybercrime group used a language model to write a Python script that bypassed two-factor authentication in a popular open-source server administration tool, and was planning a mass exploitation campaign. The vendor patched the flaw on May 11 after coordinated disclosure. The bug was a hardcoded trust assumption in the login flow. Several details gave the AI authorship away: oversized educational comments, a hallucinated CVSS score, and a clean Python structure typical of language model output.