Descriptions:
Google has significantly updated Gemini’s integration with Chrome, and the centerpiece is Auto Browse — a native browser agent that can navigate websites, fill in forms, compose emails, and complete multi-step tasks on a user’s behalf. Paul Lipsky walks through all the major changes in this hands-on overview, starting with the redesigned side panel (replacing a disruptive floating window) and a new cross-tab mode that lets users pull content from multiple open pages into a single conversation.
The Auto Browse demonstrations are the most substantive part of the video. Lipsky shows the agent navigating Etsy to add items to a cart, composing and sending an email through Gmail without switching tabs, and editing a Zillow listing photo to remove furniture using Gemini’s Imagen (referred to as Nano Banana) — all triggered from a natural language prompt in the side panel. He also shows Auto Browse drafting a YouTube script in Google Docs, using an uploaded style guide to match a specific structure and format.
Lipsky is candid about the feature’s current limitations: Auto Browse runs noticeably slower than doing tasks manually, and it frequently pauses to request user permission at sensitive steps like form submissions or address entry — meaning a task left unattended can stall mid-execution. Despite these rough edges, the update represents a meaningful step toward ambient browser automation built natively into Chrome, with Gemini serving as the orchestration layer across Gmail, Docs, Calendar, and arbitrary websites.
📺 Source: Paul J Lipsky · Published January 30, 2026
🏷️ Format: Tutorial Demo
