Google says AI fixed 1,072 security bugs in Chrome across two releases
Published On 31 Jul, 2026
The surge in discoveries is driven by Google’s aggressive integration of large language models into its vulnerability management process.
The tech giant introduced an AI-driven agent named Big Sleep, developed in collaboration with Google DeepMind and Project Zero, as well as a dedicated Gemini-powered agent to scan the codebase.
These AI systems automate nearly everything, from identifying flaws and reproducing reports to assigning severity ratings and generating candidate patches.
The automated tools are already proving highly effective. In one instance, the AI system uncovered a critical sandbox escape vulnerability concealed in the Chrome codebase for more than 13 years.
