Most People Don't Know Chrome Can Now Shop and Research For You—Hands-Free
Chrome's new Auto Browse is a full AI agent that navigates sites, fills forms, and compares prices on its own while you do something else. It's not autofill—it's autonomy.
01. What It Is
Launched on January 28, 2026, Chrome Auto Browse is an autonomous AI agent built directly into Chrome and powered by Gemini 3. Unlike traditional autofill or a simple assistant, it can open websites, click through pages, fill out forms, gather information, and compare options across multiple sites—all without you controlling the mouse or keyboard.
You give it a goal in plain language, such as researching travel packages, collecting contractor quotes, or comparing product prices, and it works through the steps on its own in the background. The whole process streams live in the Gemini side panel, so you can watch each action it takes, pause it, or click 'Take over task' whenever you want to step in.
Because it lives inside Chrome's built-in Gemini interface, there's no extension to install. It runs while you keep working in other tabs or apps, effectively turning tedious multi-site research into something that happens in the background of your day.
Why It Matters
Comparison shopping and research that used to eat 30–60 minutes of tab juggling can now run while you focus on real work. Instead of opening a dozen tabs to compare flights, gather quotes, or check prices, you delegate the whole errand and come back to a summary—saving time and often money by surfacing better deals you'd never have hunted down manually.
Who Can Benefit
- Frequent online shoppers who want the best price without endless tab-hopping
- Travelers comparing flights, hotels, and package deals across booking sites
- Homeowners collecting quotes from contractors or service providers
- Busy professionals who want to offload repetitive multi-site research
- Anyone managing subscriptions or recurring purchases across accounts
02. Step-by-Step Guide
- 1
Confirm you have the right subscription
Auto Browse requires an active Google AI Pro or AI Ultra subscription. Without one, the feature won't appear in Chrome. Note that Pro subscribers get 20 Auto Browse tasks per day, while Ultra subscribers get up to 200.
- 2
Update Chrome on a supported desktop
Use Chrome on Mac, Windows, or a Chromebook Plus device. Make sure Chrome is fully updated. The feature is currently rolling out in the US, with a phased global expansion expected afterward.
- 3
Open the built-in Gemini interface
Launch Gemini from within Chrome—no extension needed. It's integrated directly into the browser and appears in the side panel.
- 4
Describe your task in plain language
Tell the agent what you want in natural language, for example: 'Compare prices for this laptop across three major retailers and tell me the cheapest.' The clearer your goal, the better it navigates.
- 5
Let it run while you work
Auto Browse starts navigating sites, filling forms, and comparing options on its own. You can switch to other tabs or apps while it works in the background via the sidebar.
- 6
Watch, approve, or take over
Follow its real-time actions in the Gemini side panel. It will pause and ask for your approval before sensitive actions like logging in or making a purchase. You can interrupt or click 'Take over task' at any moment.
Pro Tips
- Keep tasks specific: name the exact product, sites to check, and the criterion (lowest price, fastest shipping) so the agent doesn't wander.
- Use AI Ultra if you run frequent research—200 daily tasks versus 20 makes a big difference for heavy users.
- Stay in the side panel for the first few runs so you learn how the agent behaves before trusting it with longer errands.
- When it pauses for approval on a login or purchase, that's your safety checkpoint—review carefully before confirming.
Warnings & Limitations
- Requires a paid Google AI Pro or AI Ultra subscription; there is no free tier for this feature.
- Currently limited to the US with a phased global rollout, so it may not be available in your region yet.
- Desktop only—supported on Mac, Windows, and Chromebook Plus, not on mobile.
- All browsed content streams to Google's cloud servers, so avoid using it for highly sensitive accounts or private data you don't want processed remotely.
- Daily task limits apply (20 for Pro, 200 for Ultra), so heavy sessions can be capped.
Related Tricks
You Can Turn Any Website Into a Desktop App in Under 10 Seconds
Chrome and Edge can 'install' any website as a standalone app with its own window, icon and taskbar presence — no downloads, no app store.
Your Browser Can Now Run Full AI Models Locally—With Zero Internet
WebGPU quietly shipped in Chrome, Firefox, and Safari, letting a language model run right inside a browser tab. No API key, no cloud, no subscription—and once loaded, it works with the network unplugged.
Most Google Workspace Users Don't Know NotebookLM Can Run Inside Automated Workflows
A new 'Ask NotebookLM' step turns your curated knowledge base into a hands-free workflow engine that reads, reasons, and replies while you sleep.