Explained: Google Chrome's Auto Browse and how Gemini browses the web for you
Gadgets Now January 30, 2026 04:42 AM
Google has rolled out Auto Browse, an AI-powered feature in Chrome that can navigate websites, fill forms, and complete multi-step tasks without you lifting a finger. The agentic capability, built on Google's Gemini 3 model, is now available to AI Pro and AI Ultra subscribers in the US.
Think of it as autofill on steroids. Instead of just plugging in your address at checkout, Auto Browse can actually go shopping for you. It opens a dedicated tab, scrolls through pages, clicks buttons, and enters text—all while you watch from the sidelines or do something else entirely. Gemini takes control of your browser tabWhen you trigger Auto Browse through the Gemini sidebar in Chrome, the browser spawns a new tab marked with a cursor and sparkle icon. The AI then takes the wheel. You'll see a step-by-step breakdown in the side panel showing exactly what Gemini is doing at any given moment.
The feature leans heavily on cloud processing. Your page interactions get streamed to Google's servers where Gemini handles the heavy lifting. According to Google's documentation, this activity may be logged to your account depending on your privacy settings.
For tasks requiring login credentials, Auto Browse can tap into Google Password Manager—but only after you explicitly grant permission. And it won't complete sensitive actions like final purchases or social media posts without your confirmation. 20 tasks a day for Pro, 200 for UltraGoogle has put guardrails on usage. AI Pro subscribers get up to 20 browsing tasks per day, while AI Ultra users can run up to 200 sessions daily. The company hasn't announced when free-tier access might arrive.
The use cases Google is pitching range from practical to ambitious: comparing hotel prices across multiple weekends, filling out tax documents, scheduling appointments, managing subscriptions, even getting quotes from plumbers. In one demo, an executive asked Auto Browse to reorder a jacket from last year and find a discount code before checkout. You're still on the hook for mistakesHere's the catch. You're responsible for whatever Gemini does while browsing on your behalf. That disclaimer from Google reads: "Use Gemini carefully and take control if needed. You are responsible for Gemini's actions during tasks."
Security researchers have flagged that AI browsing agents remain vulnerable to prompt injection attacks, where malicious websites can trick the bot into doing things you never intended.
Chrome will also support Google's Universal Commerce Protocol, a new standard developed with Shopify, Etsy, Wayfair, and Target that lets AI agents perform actions across participating e-commerce platforms.
The feature lands as Google plays catch-up with competitors like OpenAI's Atlas browser and Perplexity's Comet, both of which have been pushing similar agentic browsing capabilities. Whether users actually trust an algorithm to navigate the web on their behalf remains the bigger question.
