Gemini in Chrome has just been refreshed with three new features that integrate more Gemini capabilities within Chrome for Windows, MacOS, and Chromebook Plus. The update adds an AI side panel, agentic AI Auto Browse, and Nano Banana image editing of whatever image is in the browser window.
AI Side Panel For Multitasking
Chrome adds a new side panel that enables users to slide open a side panel to open up a session with Gemini without having to jump around across browser tabs. The feature is described as a way to save time by making it easier to multitask.
Google explains:
“Our testers have been using it for all sorts of things: comparing options across too-many-tabs, summarizing product reviews across different sites, and helping find time for events in even the most chaotic of calendars.”
Opt-In Requirement For AI Chat
Before enabling the side panel AI chat feature, a user must first consent to sending their URLs and browser data back to Google.
Screenshot Of Opt-In Form
Nano Banana In Chrome
Using the AI side panel, users can tell it to update and change an image in the browser window without having to do any copying, downloading, or uploading. Nano banana will change it right there in the open browser window.
Chrome Autobrowse (Agentic AI)
This feature is for subscribers of Google’s AI Pro and Ultra tiers. Autobrowse enables an agentic AI to take action on behalf of the user. It’s described as being able to researching hotel and flights and doing cost comparisons across a given range of dates, obtaining quotes for work, and checking if bills are paid.
Autobrowse is multimodal which means that it can identify items in a photo then go out and find where they can be purchased and add them to a cart, including adding any relevant discount codes. If given permission, the AI agent can also access passwords and log in to online stores and services.
Adds More Features To Existing Ones
Google announced on January 12, 2026 that Chrome’s AI was upgraded with app connections, able to connect to Calendar, Gmail,Google Shopping, Google Flights, Maps, and YouTube. This is part of Google’s Personal Intelligence initiative, which it said is Google’s first step toward a more personalized AI assistant.
Personalization And User Intent Extraction For AI Chat And Agents
On a related note, Google recently published a research paper that shows how an on-device and in-browser AI can extract a user’s intent so as to provide better personalized and proactive responses, pointing to how on-device AI may be used in the near future. Read Google’s New User Intent Extraction Method.
Featured Image by Shutterstock/f11photo
