Google Uses AI to Fix 1,072 Chrome Security Vulnerabilities

Google utilized artificial intelligence to identify and patch 1,072 security vulnerabilities across Chrome 149 and Chrome 150, exceeding the total fixes from the previous 23 browser releases.

The AI-driven process, powered by Gemini-based agents, successfully discovered a 13-year-old sandbox escape flaw that had previously evaded conventional testing and human review.

While AI agents now automate vulnerability discovery and patch generation, human developers retain final responsibility for reviewing and approving all code changes before production.