< Back to all clusters
[TECHNOLOGY] · United States · 13 sources

started · updated

Google Chrome AI agents enable record security fixes and dynamic patching

Google has integrated Gemini‑powered AI agents into Chrome’s security workflow, dramatically increasing the pace of vulnerability discovery and remediation. The agents identified a 13‑year‑old sandbox‑escape flaw (CVE‑2026‑3545) and helped fix a total of 1,072 security bugs across milestones 149 and 150, a count that exceeds the combined total of the previous 23 releases.

The company is piloting a twice‑weekly update cadence and developing “dynamic patching,” a technique that uses Chrome’s multi‑process architecture to replace background child processes (such as the Renderer and GPU) on the fly, eliminating most user‑initiated restarts. A first implementation appears in Chrome 150 on macOS, where the browser can restart automatically when no windows are open.

Google also announced a shift to a two‑week release cycle for Chrome, with Chrome 153 slated to launch the new cadence in September 2026. Subsequent updates, such as Chrome 151, patched 370 additional flaws, including seven rated critical. The rapid release cadence has already caused stability issues for some Mac users, who reported crashes and freezes after the large batch of security patches.

Entities

BigSleep · CVE‑2026‑3545 · Chrome · Chrome 150 · Chrome 151 · Chrome web browser · CodeMender · Dynamic Patching · Gemini · Gemini (Google AI model) · Gemini AI model · Google

Claims

What the coverage asserts, and how many sources carry each claim.

Sources

about 2 months ago