[The Age of AI Assistants] ④How AI Tab Is Transforming Search: From Restaurant Reservations to Product Purchases
Integration with Naver's Ecosystem, Including Shopping and Place
Immediate Reservations for Recommended Restaurants... Instant Purchase of Suggested Products
Personalized Recommendations Available... Quick Responses to the Latest Trends
Slow Response Times and Lack of Direct Execution Remain Drawbacks
Will there come a time when artificial intelligence (AI) can replace our work? This was once a distant future scenario, fit for movies, but with the emergence of AI agents, it is becoming reality. The defining feature of AI agents is that they make decisions on behalf of users, plan, and automatically carry out tasks. We personally tried out several AI agent services that companies have launched in succession and analyzed how conveniently they can be applied to our daily lives.
"If you want to enjoy a glass of wine on a weekday evening on Seosunragil and are considering reservations, restaurants A, B, and C are the most reliable options. As of today, all three have available reservation slots, and each features a different ambiance and cuisine, so you can choose according to your preferences. Would you like me to check the menu selection or price range for these recommended restaurants?"
The answer provided by Naver's AI Tab when asked, "Please recommend a good restaurant where one can enjoy a glass of wine on a weekday evening on Seosunra Road, and where reservations are possible." Screenshot.
View original image
When a request was made to Naver AI Tab to "recommend a good restaurant on Seosunragil where it's nice to have a glass of wine on a weekday evening and reservations are possible," this was the response. Below the answer, a table summarized each restaurant's features, reasons for recommendation, and available reservation times. Clicking on a restaurant name opened a small window displaying details such as business hours and menu from Naver Place. Right next to the restaurant information, a "Naver Reservation" button was activated.
Naver began the beta service of its AI Tab, an artificial intelligence (AI)-based search service, on April 27. In AI Tab, users can interact naturally through conversation, similar to using an AI chatbot. Currently, the AI Tab feature is offered to paying Naver Plus Membership subscribers.
Pinpoint Recommendations by Collecting Restaurant Reviews... One-Stop Shopping and Reservations
Users can access AI Tab by clicking the "AI" button next to the search bar on the Naver PC version or the "Learn More with AI" button in the middle of mobile search results. The main screen of AI Tab resembled that of an AI chatbot like ChatGPT or Gemini rather than a typical search bar. There was a field for inputting questions, and beneath it, sample questions were displayed. Through the settings button below the chat window, users could set up specialized agents for services such as "shopping" and "restaurant search."
After actually using AI Tab, we found that it provided useful recommendations, especially in categories where Naver excels, such as shopping and Naver Place. For example, when asked, "Please recommend a laptop with adequate performance and light weight suitable for a college student nephew. The price should be around the mid-1 million won range," the AI Tab presented a list of recommended models. It introduced products based on various criteria such as performance, portability, and price, and also explained how components like the CPU and RAM affected performance. Sources of information were clearly indicated, including posts by well-known IT bloggers and product details from the Naver Plus Store.
The result of a request to Naver AI Tab saying, "Please recommend a laptop with adequate performance and lightweight suitable for a college student nephew. The price should be around mid-1,000,000 KRW." Screenshot.
View original image
Naver's AI Tab even incorporates some features of an AI agent. Unlike conventional chatbots, AI Tab allows users to make actual purchases and reservations directly within the platform. Along with introductions to the recommended laptops, product information from the Naver Plus Store—where the items could be purchased—was presented, and clicking this brought up a store window next to the chat. Similarly, when restaurants or cafes were recommended, relevant Naver Place (map) information was displayed. If a business supported Naver Reservation, users could make reservations as well.
The journey from casual conversation to product search and purchase felt seamless through interactions with AI Tab. For instance, when we asked why Chang-eok Tteok is trending, the AI Tab provided a summary of how it became popular online, blog reviews introducing the product, and an overview of the company behind it. The AI Tab then offered to provide information on where to buy Chang-eok Tteok and its prices; upon request, it presented the price per weight unit at each seller and a link to the Naver Plus Store page, enabling purchase with just a few clicks.
The system also offered responses based on personalized context such as purchase history. For example, when asked, "Please recommend a supplement that pairs well with the red ginseng I bought last month," it analyzed the user’s purchase history and suggested suitable supplements. Users could even directly inquire about past purchases through Naver Plus Store, such as "Which brand was the cosmetic I bought recently?"
It was also helpful for searching practical information or checking recent trends. When asked "When and how much can I receive the basic livelihood support fund?", eligibility, conditions, application period, and calculation criteria were neatly summarized in a table. When questioned about the short-form video-based social networking service "Setlog," which is popular among Generation Z, the AI Tab provided a detailed introduction to the app, reasons for its popularity, and instructions on how to use it.
However, as this is still a beta service, there were some incomplete aspects. In some cases, responses took several tens of seconds to generate. While it could recommend products or locations matching the request, it does not yet support purchasing or making reservations solely through conversation. Currently, users must still click the "purchase" or "reserve" button themselves and complete payment or reservation directly.
Hot Picks Today
"Semiconductor Peak-Out Theory Resurfaces: Foreign Investors' KRW 3.2 Trillion Net Sell Pushes KOSPI Below 8,000"
- [Exclusive] "Tens of Billions Paid": Unregulated Bonuses for Union Presidents Spark Controversy... Ministry of Land Launches Nationwide Investigation
- "Now Is Not the Time to Sell, Buy More on Sharp Declines"... Advice Amid Market Volatility
- "Why Did You Call Me to the World Cup!" Jens-Hong Myung-bo Assault Video Surpasses 10 Million Views... Revealed as AI
- Suspicious Achievements of Famous 'Beauty' Policewoman Revealed to Be Handed Over by Male Colleagues... Taiwan in Shock
Naver plans to optimize response speed and enhance performance to deliver more sophisticated answers to complex, continuous queries by the official launch of AI Tab in the first half of this year. Within the year, Naver aims to strengthen the multimodal AI search experience by linking AI Tab with SmartLens and to upgrade the service so that AI Tab can proactively suggest follow-up questions and explorations to users.
© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.