What to know
-
Gemini in Chrome is getting a new side panel designed for multitasking across tabs.
-
A new “auto browse” mode can carry out multi-step web tasks on your behalf (with confirmations for sensitive actions).
-
Chrome is adding Connected Apps support for services like Gmail, Calendar, Maps, and Flights.
-
Google says Chrome will support Universal Commerce Protocol (UCP) to help AI agents take commerce actions across sites.
Google has unveiled major updates to Gemini in Chrome for macOS, Windows, and Chromebook Plus, built on the Gemini 3 model. The changes aim to make Chrome feel more like an assistant that can both help find information and take action across the web.
A new Gemini side panel for multitasking
Gemini in Chrome is moving into a new side panel so assistance stays available “no matter what tab” is open. This setup is positioned for tasks like comparing options across many tabs, summarizing product reviews across sites, and helping coordinate calendars without leaving the main page.
“Auto browse” brings agentic actions to Chrome
Chrome is adding “auto browse,” an agentic feature that can handle multi-step chores such as researching travel options across dates, filling out forms, collecting documents, getting quotes for services, filing expense reports, and managing subscriptions. Google also describes multimodal use cases, such as identifying items from an inspiration image and adding similar products to a cart while staying within budget and applying discount codes.
Auto browse is launching for AI Pro and AI Ultra subscribers in the U.S.
Nano Banana image edits inside the browser
Chrome is also bringing Nano Banana image transformation into Gemini in Chrome, allowing image edits without downloading and re-uploading files or opening another tab. This capability is available to all Gemini in Chrome users via prompts in the side panel.
Connected Apps and Personal Intelligence
Gemini in Chrome will support Connected Apps integrations including Gmail, Calendar, YouTube, Maps, Google Shopping, and Google Flights, and these can be enabled in the Connected Apps section of Gemini Settings.
“Personal Intelligence” is coming to Chrome in the coming months, with opt-in controls and the ability to connect or disconnect apps; it’s intended to remember context from past conversations for more tailored help.
Commerce and safety guardrails
Chrome will support Universal Commerce Protocol (UCP), an open standard for agentic commerce co-developed with partners including Shopify, Etsy, Wayfair, and Target, aimed at helping AI agents take actions on a user’s behalf more seamlessly. For safety and control, Google says auto browse is designed to pause and request confirmation for sensitive actions such as purchases or social posts.
Chrome is one of the most widely used browsers, so embedding agent-like AI features directly into the browsing interface could speed up mainstream adoption of “agentic” workflows. Limiting auto browse to paid U.S. subscribers (at least initially) suggests Google is treating autonomous actions as a premium, higher-risk feature that needs staged deployment.
