Google AI Supercharges Chrome Security, Fixing 1,072 Bugs

Google reports that AI-driven tooling (e.g., Gemini-based agents) identified and helped fix 1,072 Chrome security bugs across two releases, including a 13-year-old sandbox-escape; the piece describes automated triage, patch generation and testing, CI integration, efforts to reduce the patch gap (dynamic patching), and long-term defenses such as MiraclePtr, compiler-based spatial checks, and migrating high-risk components to Rust.