The Biggest Google Search Redesign Since the Original Box
Google is overhauling its classic search box with what executives describe as the biggest upgrade in more than 25 years. Search remains central to Google’s business, and the company is reshaping it around Google Search AI rather than treating AI as an optional side mode. Instead of forcing people to choose between a traditional list of blue links and an AI chatbot, Google is blending the two experiences. Results pages will still show web links and AI summaries, but they are now framed by a more dynamic, AI-aware interface that can expand for complex questions and support conversational follow‑ups. By closing the gap between AI Overviews, AI Mode, and the search results page, Google aims to reduce friction so users no longer have to think about which experience to pick — they just start typing in the familiar box and let the system adapt.
Inside the Multimodal Search Box: Text, Images, Files, and Tabs
At the heart of the Google Search redesign is a multimodal search box powered by Google’s Gemini 3.5 Flash model. Instead of treating text queries, image searches, and document uploads as separate workflows, the new interface accepts text, images, files, video, and even Chrome tabs in one place. Users can, for example, drop in a PDF lease, a screenshot of an apartment listing, or an open browser tab and ask questions across all of them at once. As you type, the box offers AI-powered suggestions that go beyond simple autocomplete, surfacing richer ideas and follow‑up prompts. This multimodal search approach is designed to capture the way people actually research today, moving between formats and sources. It also tightens the relationship between the core search page and AI Mode, letting users smoothly transition from a one‑off query into a full conversational session.
Information Agents: Always‑On Search for Google AI Pro and Ultra
On top of the visual Google Search redesign, Google is introducing Google information agents that continuously search on a user’s behalf. These agents, rolling out first to Google AI Pro and Ultra subscribers, act like autonomous research assistants running in the background. Users define a goal — such as finding a specific pair of shoes in their size or monitoring apartments that match a detailed set of requirements — and the agent then plans how to search, which tools to use, and which sites to track. It scans blogs, news sites, social media, shopping platforms, and real‑time feeds like finance or sports, then sends synthesized updates when something relevant changes. Instead of repeatedly issuing queries, subscribers offload the task to agents that watch the web continuously, blurring the line between traditional search and automated, persistent monitoring of information.
Unifying Gemini, AI Mode, and Personal Intelligence Around Search
These changes reflect Google’s broader strategy to unify its AI offerings around the search experience instead of scattering them across separate apps and modes. AI Mode, which already behaves like the Gemini chatbot with follow‑up questions and long‑form answers, is being woven directly into Search so that an AI Overview can fluidly turn into a conversation. Google is also expanding Personal Intelligence, which pulls from services like Gmail and Photos to give more tailored responses, into more languages and markets via the Gemini app, AI Mode in Search, and the Gemini side panel in Chrome. With AI Mode reportedly reaching over 1 billion monthly users, Google is betting that most people should not have to decide whether they want a chatbot or classic results. The multimodal search box and information agents together signal that Search itself is becoming the primary interface to Google’s entire AI ecosystem.