Chrome adds agentic browsing with Gemini, including in-context image transformation via Nano Banana
Jan 29, 2026
Key Points
- Google embeds agentic AI directly into Chrome with Gemini's 'auto browse' feature, automating routine web navigation tasks within the browser rather than requiring dedicated AI applications.
- Nano Banana, a new image transformation model, edits photos in-context on webpages, eliminating the friction of downloading, editing separately, and re-uploading files.
- Google's strategy embeds AI agents into surfaces where users already spend time, moving beyond text generation toward direct manipulation of on-screen content.
Summary
Google launched agentic browsing capabilities in Chrome, adding an 'auto browse' feature that automates routine web navigation tasks and a new image transformation model called Nano Banana. Nano Banana operates directly on images visible on screen, processing and modifying them in-context without requiring users to download, edit separately, and re-upload them.
Addy Osmani, a Google Cloud engineer, announced the updates as part of a broader set of Chrome and Gemini changes. The capability embeds AI agents deeper into Google's browser and product stack, moving beyond text generation into direct manipulation of on-screen content. Google's strategy prioritizes integrating agentic AI into surfaces where users already spend time rather than forcing them into dedicated AI applications.