Google is turning its Chrome browser into a full-fledged AI agent. With the launch of Gemini 3, Chrome now features “Auto Browse,” a tool that can perform multi-step tasks like booking flights, filing expenses, and even shopping on your behalf.
In a major move to solidify its lead in the AI browser wars, Google has announced the integration of its latest Gemini 3 model into Chrome. This update moves beyond simple chatbots, introducing “agentic” capabilities that allow the browser to take action across the web without the user needing to click every button themselves.
The Rise of “Auto Browse”: Your Browser is Now an Agent
The headline feature of this update is Auto Browse. While Chrome has long handled simple tasks like autofilling addresses, Auto Browse is designed for “agentic action.” This means the AI can handle complex, multi-step workflows that usually require jumping between multiple tabs and forms.
According to Parisa Tabriz, VP of Chrome, the feature is currently rolling out in preview for Google AI Pro and Ultra subscribers in the U.S. Use cases for Auto Browse include:
- Travel Planning: Researching hotel and flight costs across various dates and presenting the best budget options.
- Administrative Chores: Filling out tedious online forms, collecting tax documents, and even managing subscription cancellations.
- Agentic Commerce: Chrome can now identify items in a photo, find them for sale, apply discount codes, and add them to your cart.
To support this, Google is co-developing the Universal Commerce Protocol (UCP)‘an open standard with partners like Shopify, Etsy, and Wayfair’to ensure AI agents can navigate checkouts seamlessly.
Multitasking with the New Side Panel
Google is also introducing a persistent Gemini side panel for Windows, macOS, and Chromebook Plus. This allows users to keep their main work in the center tab while Gemini assists in the periphery. The side panel can summarize long articles, compare product reviews across tabs, or cross-reference your Google Calendar to find meeting times without leaving the page.
Nano Banana: On-the-Fly Image Generation
Chrome is also integrating Nano Banana, Google’s latest creative AI tool. Users can now generate or transform images directly within the browser side panel. For example, a user can prompt the AI to turn raw research data into an infographic or visualize a room redesign, all without having to download files or open separate editing software.
Personal Intelligence and Ecosystem Integration
In the coming months, Google will bring “Personal Intelligence” to Chrome. This opt-in feature allows the browser to remember context from past conversations and user preferences to provide tailored answers.
Crucially, Chrome is deepening its integration with Connected Apps. Gemini can now “reach into” Gmail, Maps, and Google Flights. For example, it can find an old email about a conference, check flight availability, and draft a notification to colleagues about your arrival time in one go.
Security: Keeping the Human in the Loop
Recognizing the risks of “agentic” AI, Google has built-in safeguards. Auto Browse is designed to pause and ask for explicit user confirmation before high-stakes actions, such as making a purchase or posting to social media. The browser also utilizes the Google Password Manager to handle logins securely when the AI needs to access password-protected accounts to complete a task.
Availability
- Gemini Side Panel: Rolling out now for Windows, macOS, and Chromebook Plus users in the U.S.
- Auto Browse: Available in preview for Google AI Pro and Ultra subscribers.
- Personal Intelligence: Expected to roll out in the coming months.