- Google Search now uses Gemini 3 to generate responses to AI insights
- The upgrade aims to improve answers to more complex questions
- The update also allows users to go directly from AI previews to an AI mode conversation.
Google Search has upgraded its AI Overview Answers with its Gemini 3 AI model to better answer complex queries in a single snapshot. The update also lets you switch from an AI presentation response directly to a Google Search AI mode conversation on mobile without having to switch tabs or start over.
AI Previews, those small summaries that sit on top of traditional search results, are designed to condense long, messy answers to complex questions without requiring you to click on links. Making Gemini 3 the default for mobile AI previews globally means responses will be smarter, longer and better structured, according to Google.
As improved as these answers are, they may not be everything you want. This is where the connection between static AI previews and interactive AI mode for Google Search comes into play. The upgrade makes it easier to move from that snapshot to a full conversational exchange with AI mode.
Previously, if you wanted to turn your preview into a full conversation, you had to click on the AI Mode tab and start searching again. Now you can stay in the flow of your search results and continue digging deeper with minimal friction. In early testing, Google found that people naturally wanted these conversational search extensions, rather than hitting a wall after the first AI-generated response.
While the overview gives you a starting point, the conversation lets you discover every detail without having to start a new search. You can then refine your question, ask about exceptions, or even integrate tables and other visuals within the same discussion thread.
Conversational transitions
Gemini 3 makes both quick snapshot and deeper conversation more useful. Google hopes that if your exploration of a topic doesn’t stop after the first answer, you can satisfy your curiosity without leaving its sandbox.
For most people, this means that the understandable summary at the top can immediately become a deeper, more nuanced conversation without switching tabs, rewriting your question, or losing context. It’s a search that remembers what you ask and anticipates follow-ups.
Context matters here. Research has always been about finding quick answers to simple questions like sports scores or the weather, but life’s big questions are often complex and nuanced. Knowing that “what to consider before buying a house” isn’t something you solve with a snippet, Google’s conversational layer seems timely. The ability to request follow-ups and maintain context means fewer redundant queries and less reframing for more subtle angles on your issue.
Of course, not all questions will be best served by conversation. Google’s efforts to use the right model where appropriate, with lighter models for quick answers and Gemini 3 for deeper queries, reflect this balance. And there is a question of accuracy. No matter how well AI models perform, hallucinations arise, and well-crafted responses and dialogues aren’t very useful if they’re flat-out wrong.
But a search experience that feels less like fishing for answers and more like a fish dinner prepared based on your conversation with the chef will undeniably please those in a hurry, even if the fish is sometimes made of rubber.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds. Make sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp Also.
The best business laptops for every budget
