NAVER Map Launches Place Agent, Enabling One-Stop Place Discovery and Booking Through Everyday Conversation
NAVER Map Launches Place Agent, Enabling One-Stop Place Discovery and Booking Through Everyday Conversation
NAVER Map Launches Place Agent, Enabling One-Stop Place Discovery and Booking Through Everyday Conversation
- AI-powered conversational search designed for place discovery recommends spots based on conversation context and the user’s location, and makes it easy to go straight to booking
- Stands out for finding suitable places even when a single request packs in multiple conditions, such as “Korean restaurant in Yongin with plenty of parking, good for a stop on the way back from visiting ancestral graves,” and draws on rich data to explain each place’s features and why it was recommended
- Also lets users request reservations easily by providing all the necessary details, from date and time to party size, in a single sentence
September 17, 2026
NAVER Corp. (CEO Choi Soo-yeon) has unveiled Place Agent, a conversational place discovery solution for the NAVER Map app that lets users get recommendations and make reservations in one go, simply by chatting in everyday language.
Place Agent is an AI-powered conversational search feature designed specifically for finding places. Within the NAVER Map app, it recommends places based on the context of the conversation and the user’s location, then takes users all the way through to booking in one seamless flow.
Drawing on the wealth of place information on NAVER Map, from restaurants, cafes, and bakeries to hotels, guesthouses, hair salons, and veterinary clinics, the agent recommends the places that best match what users are looking for and connects users to them directly.
Users can launch Place Agent by tapping the Agent button, which appears to the right of the search bar when they select it in the NAVER Map app. They can also reach the agent by tapping one of the suggested questions shown among restaurant search results, or via the “Ask AI Agent” button that appears when no matching results are found.
Beyond understanding conversational, full-sentence queries, Place Agent is also strong at handling multiple search conditions in a single sentence, taking all of them into account to find places that fit.
For example, users can enter specific requests such as “Recommend a bakery-hopping route in Seongsu-dong,” “Quiet cafe in Yeonnam-dong for reading alone on a rainy day,” “Great tofu place in Ui-dong for after a fall foliage hike on Bukhansan,” or “Korean restaurant in Yongin with plenty of parking, good for a stop on the way back from visiting ancestral graves,” and Place Agent then presents a list of the places that best meet those conditions.
Recommendations need not be tied to a specific area. Place Agent can also suggest places based on the situation users describe, such as “a safe bet for a date” or “somewhere with easy parking.”
Until now, finding a place in a map app meant searching several times with separate keywords, such as exact business names or menu items, and then manually comparing information across the resulting place listings. Place Agent is expected to make it easier for users to find the places they want.
Notably, Place Agent’s answers draw on a wealth of useful information held by NAVER, including place details such as addresses, business hours, menus, and phone numbers, along with visitor reviews, blog reviews, and booking availability. Using this information, the agent supports each suggestion by highlighting the place’s strengths and explaining why it was recommended, helping users make sound decisions.
The agent also supports multi-turn dialogue, allowing users to continue the conversation and refine their criteria based on its answers. For instance, users can narrow down a list of suggested places by following up with questions like “Which of these are quiet?”
For places that use NAVER Booking, users can also request a reservation by giving the agent all the necessary details, including date, time, and party size, in a single sentence.
When users enter a request such as “Book a table for four at [business name] for lunch tomorrow at noon,” the agent replies with a summary of the reservation details and asks them to confirm. All they need to do is accept.
The newly launched Place Agent is available to logged-in users aged 14 and older, and the agent’s answers always carry a separate notice indicating that they are AI-generated.
“Place Agent brings together our extensive place information to deliver a service optimized for the map environment,” said Choi Ji-hoon, Leader of Place Search at NAVER. “We will continue to enhance its features and usability so that users can discover the places they want more easily and accurately, and move naturally from discovery to booking and an actual visit.” (End)
