Google is overhauling Chrome security by using Gemini-powered AI tools, alongside systems from Google DeepMind and Project Zero, to find, sort, and fix bugs faster. Across the Chrome 149 and 150 updates, Google fixed 1,072 security bugs, which is more than all the bugs fixed in the previous 23 updates put together. The AI system found a serious flaw that had been hiding in Chrome's code for over 13 years. By March 2026, bug reports had already crossed the total number seen in all of 2025. Google built an automated system where "fixing agents" write possible code fixes, "critic agents" check whether fixes meet standards, and "test-writing agents" create tests. These tools check every code change every 24 hours, and in May alone, they stopped more than 20 bugs, including a very serious one, from reaching users. Google is also addressing memory safety, which causes many serious browser bugs, with a long-term plan to shift to safer coding languages like Rust while still fixing older C++ code. To shorten the "patch gap" after fixes become public, Google is testing two security updates a week instead of one. Additionally, "dynamic patching" lets Chrome quietly replace background parts, like the Renderer and GPU, without needing a full restart.