Google brings Gemini-powered conversational search and reporting to Waze

Google rolled out a set of Gemini-powered features to Waze on July 13, letting drivers search for destinations and report road conditions by talking naturally to the app instead of typing or picking from menus, alongside a new AI-driven motorcycle routing mode.

What's new

The headline addition is conversational destination search: drivers can tap a voice icon and ask open-ended questions like "find me a coffee shop that's open right now" or "find me parking close to Grand Mall," rather than typing an exact address or business name. Waze already used Gemini for Conversational Reporting — letting users flag a slowdown or hazard by speaking — and Google is extending that same capability to map corrections: "Just say, 'The road is closed here' and Waze will send these details to local map editors, who verify the suggestion and update the map."

Routing itself is also getting smarter about individual habits — Waze will now factor in a driver's past trip choices, not just current traffic, so a driver who consistently favors surface streets over highways will see those routes prioritized. A new "less chatty" mode trims the frequency of voice prompts to essentials like hazards, turns, and lane changes for drivers who want less audio interruption.

The update also adds a dedicated motorcycle mode: "Our new motorcycle mode uses AI to incorporate two-wheeler shortcuts and restrictions, helping you find the best route and get an accurate ETA," accounting for road surfaces, narrow bridges, and hazards like potholes and raised crosswalks that matter more for two-wheeled vehicles than cars.

Context

Waze has folded Gemini into its product incrementally since Google began pushing the model family across its consumer app portfolio — Search, Maps, Assistant, and now Waze all lean on the same underlying model rather than bespoke NLP systems per app. Conversational search and reporting specifically extend a pattern Google has used elsewhere: replacing structured input (typed queries, menu taps) with open-ended natural language wherever an app already has a voice interface.

Why it matters

Navigation apps are a high-frequency, safety-sensitive surface — drivers can't look at a screen while driving, which makes voice-first interaction more than a convenience feature. Getting conversational search right in a moving vehicle, where background noise and short attention windows are the norm, is a harder product problem than the same feature in a chat window, and a meaningful proving ground for Gemini's real-world reliability outside controlled settings. The habit-aware routing and less-chatty mode also point to a broader shift: personalization and reduced friction, not just raw capability, increasingly differentiate one AI-powered consumer feature from another once most competitors have access to similarly capable underlying models.