Google bakes Gemini AI into Chrome on Android, turning the browser into your on‑device assistant

Google is rolling out a deeper integration of its Gemini AI assistant directly into Chrome on Android for users in the United States, and tech media are treating it as a glimpse of what “AI browsers” will look like. From the Chrome side panel, Android users can now invoke Gemini to summarize the page they’re viewing and ask follow‑up questions about the content, turning long news articles or blog posts into quick briefings.

For paying AI Pro and Ultra subscribers, Google is adding an even more ambitious capability called auto‑browse. Instead of just answering questions, Gemini can carry out multi‑step tasks inside the browser — like booking an event or placing an online order. You type a prompt, Gemini generates a step‑by‑step plan, and only after you review and approve it does the assistant execute those actions in Chrome.

There’s also a playful experimental feature nicknamed Nano Banana that lets users generate images directly inside the browser, useful for quick sketches or social thumbnails. Google emphasizes an on‑device–first design, reducing the need to send every interaction to remote servers. While there’s no timeline yet for a Japan rollout, the update is being watched closely as an early example of the mobile browser itself becoming the primary interface to AI services.