Google Maps Adds Agentic Ordering and Personal Intelligence
Google expanded Ask Maps on August 6 with agentic food ordering, hotel search, event-ticket discovery, and Gemini Personal Intelligence. The Verge reports that Ask Maps can add a selected meal to a cart through food-ordering partners, while TechCrunch reports users complete payment on the supported platform. Personal Intelligence can draw on Gmail reservations when a user manually enables it, according to 9to5Google.
Google has expanded its Gemini-powered Ask Maps feature in Google Maps with agentic food ordering, hotel search, event-ticket discovery, Gmail-based personalization, conversation memory, and real-time transit information.
The food-ordering workflow can interpret requests that combine meal preferences, location, saved places, and a route. The Verge reports that, after a user selects a restaurant, Ask Maps can add a selected dish to a cart for the user to complete the order. Google Maps head Miriam Daniel said the updates are intended to "make it easier than ever to get more done" without leaving the Maps app, according to The Verge.
TechCrunch reports that users can select "Order online" after Ask Maps identifies a restaurant, add further items to the cart, and pay through the supported ordering platform. The Verge reports that Square and Toast support is rolling out now, with Uber Eats support coming soon.
Broader task workflows
The release extends Ask Maps beyond restaurant discovery. According to TechCrunch, users can ask the tool to compare hotel prices and availability against constraints such as location, price, ratings, and nearby amenities, then click through to a partner website to book. Ask Maps can also return local comedy-show and live-music options with links to buy tickets, TechCrunch reports.
These workflows remain assisted transactions rather than fully autonomous purchases. The reported food-order flow stops at a cart, and the hotel flow directs users to a partner site for booking. That design preserves a user confirmation and payment step while allowing the conversational interface to handle search, filtering, and item selection.
Personal context and transit data
9to5Google reports that Ask Maps can use Gemini Personal Intelligence, beginning with Gmail, after users manually enable the setting. The integration can use information such as flight or dinner reservations to tailor local recommendations around existing plans. Google Calendar and other app integrations are described by 9to5Google as coming soon.
Ask Maps can also recall previous conversations when History is enabled, according to 9to5Google. In addition, the feature now surfaces real-time transit information for buses, trains, subways, and ferries, including delay updates in a chat-based departure-board view.
9to5Google reports that Ask Maps is expanding in English to Australia, Brazil, Canada, Indonesia, Japan, Mexico, and more than 150 additional countries and territories. It also reports that the Personal Intelligence, conversation-memory, and real-time-transit features are rolling out wherever Ask Maps is available.
For AI product teams, the update illustrates a common agentic-interface pattern: combine conversational retrieval with user-authorized personal context and external commerce or booking systems, while retaining explicit handoffs for payment and final transactions. The practical quality of such systems depends heavily on live partner inventory, reliable tool execution, and clear consent controls for connected personal data.
Key Points
- 1Ask Maps now combines conversational restaurant search with cart creation, while payment remains on a supported third-party ordering platform.
- 2Gmail-based Personal Intelligence can tailor Maps recommendations around reservations, but users must manually enable the connected-data feature.
- 3Comparable agentic interfaces depend on dependable partner inventory, tool execution, user consent, and clear transaction-confirmation boundaries.
Scoring Rationale
The update brings agentic task completion and personal-context retrieval to a widely used consumer mapping product. It is relevant to practitioners building tool-using assistants, particularly around transactional handoffs, connected-data consent, and real-time retrieval, though it is not a new foundational model or developer platform.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
