The capability, called “auto browse,” reflects a shift in how people might interact with online content, moving from manual search and navigation to delegating complex workflows to an AI assistant.
Auto browse is powered by Google’s Gemini 3 AI model and is initially available in preview for Google AI Pro and AI Ultra subscribers in the United States on desktop versions of Chrome. It allows users to describe goals in natural language, such as researching travel options, comparing products or filling out forms. Chrome’s AI will open sites, click links and work across pages to complete those tasks. The rollout represents one of the most advanced uses of generative AI directly inside a mainstream browser.
According to Bloomberg, this new agentic capability can interact with websites on the user’s behalf, reducing the need for manual navigation and repetitive input. Google has said that sensitive actions like making purchases or posting on social media will still require explicit user approval before final execution, aiming to balance convenience with user control.
The feature builds on earlier additions of AI into Chrome, where the browser could already summarize page content, provide natural-language answers and consolidate information across tabs. With auto browse, those assistive functions evolve into proactive action: the AI doesn’t just help find information; it can execute sequences of actions across websites to fulfill user requests.
In its announcement, Google positioned auto browse as part of its broader effort to transform Chrome into an AI-native productivity platform that works alongside users’ workflows. A newly redesigned AI side panel remains open alongside tabs so users can monitor, tweak or redirect the agent’s tasks in real time. Google described the experience as akin to having a personal assistant that can operate within the same web interface that users are on.
Advertisement: Scroll to Continue
Competitors have already begun introducing their own AI-driven browsers, including OpenAI’s ChatGPT Atlas and Perplexity’s Comet, which also offer AI agents capable of navigating the web and performing actions for users. These developments reflect a growing AI browser arms race in which multiple tech companies seek to redefine how people access and use the internet.
