Google begins rolling out Chrome's "Auto Browse" AI agent today
The Auto Browse agent is available to AI Pro and AI Ultra subscribers, but there are some limits.
TL;DR
- Gemini in Chrome now features an autonomous browsing agent named Auto Browse.
- Auto Browse can perform tasks previously done manually in a browser, such as filling forms or copying information.
- Gemini is now more accessible from the top of the Chrome window, defaulting to a split-screen view.
- Gemini can now connect to and interact with services like Gmail, Calendar, YouTube, Maps, Google Shopping, and Google Flights.
- The update includes image editing capabilities within Chrome using Gemini.
- Auto Browse is based on Gemini 3 models and the experimental Project Mariner agent.
- Tasks performed by Auto Browse are marked with an AI icon, and multiple tasks can run concurrently.
- AI Pro subscribers get 20 Auto Browse tasks per day, while AI Ultra subscribers get 200.
- Auto Browse has safety measures to prevent misuse, such as not completing purchases automatically.
- Auto Browse operates via cloud-based Gemini models, with page content temporarily logged to a Google Account.
- The feature is currently in preview and exclusive to AI Pro and AI Ultra subscribers.