Google Enhances Chrome's AI Security with New Defense Layer

Google is adding a new defense layer called 'User Alignment Critic' to Chrome to enhance security for its Gemini AI agentic browsing features. This new layer is designed to protect users as the AI autonomously navigates and interacts with web content. The introduction of this feature follows the initial integration of a Gemini-powered chat window in September 2025, which raised security concerns due to the potential risks of error-prone AI models operating without human oversight.