Google Agentic AI Finds Over 100 Critical Security Flaws in Just Two Days
Brief
Google Threat Intelligence Group has revealed that its Agentic Vulnerability Discovery Harness (AVDH) identified more than 100 true-positive critical vulnerabilities in a stolen corporate code repository within two days.
The findings arrive as adversaries increasingly use AI-assisted capabilities to exploit exposed source code at machine speed. When proprietary repositories are stolen, organizations must rapidly determine which flaws are exploitable, where attack paths exist, and which systems require immediate remediation.
Google said its AVDH framework is designed to help defenders move from broad code visibility to prioritized, human-validated findings at scale.
Google Agentic AI Finds Over 100 Critical Security Flaws
AVDH combines large language models with a structured, deterministic orchestration layer based on Google’s Agent Development Kit.