Key Takeaways
- KAYAK, the Booking Holdings-owned travel metasearch giant, released Ask AI, a conversational tool that updates flight, hotel and car rental results in real time on the same screen while travelers plan in chat
- Behavioral data from more than 100,000 monthly AI conversations showed that AI works at the exploratory start of planning while travelers revert to traditional search and filters to decide, leading to a design where chat runs alongside search rather than replacing it
- As the race to make conversational AI the entry point for bookings accelerates, the insight that only suggestions grounded in live inventory and prices earn trust offers practical guidance for travel and commerce operators designing AI-assisted experiences
Ask AI: Search Results That Keep Moving Next to the Chat
KAYAK: How behavioral data from 100,000+ monthly AI conversations shaped Ask AI, which combines conversational planning with live search
www.cxtoday.comKAYAK, one of the largest travel metasearch platforms, announced Ask AI on April 23, 2026, a conversational travel planning tool. As travelers describe what they want in plain language, the flight, hotel and rental car results on the same screen update in real time as the chat progresses. There are no redirects to another page and no restarting the search every time conditions change.
The timing coincides with the World Cup being held across three North American countries. KAYAK's data shows flight searches to U.S. host cities up 23% year over year, with Kansas City surging 168%. Hotel prices have climbed 36% in the U.S., 55% in Canada and 119% in Mexico, and demand for complex multi-city itineraries is spiking. Finding a hotel near a stadium, comparing flights between host cities and building an itinerary that includes a rental car, all within a single conversation — Ask AI was launched as the answer to that complexity.
What 100,000 Monthly Conversations Revealed: The Chat-to-Search Disconnect
Ask AI's design is not an off-the-cuff feature addition. Its starting point is behavioral data from AI Mode, the natural language search KAYAK introduced in October 2025. AI Mode combined ChatGPT with KAYAK's travel data and was the first feature to graduate from the company's internal sandbox, KAYAK.ai, to the main site. Typing "family-friendly hotels in Paris under $200" into the search box returns bookable options drawn from hundreds of providers.
A clear pattern emerged from running it. According to Chief Product Officer Matthias Keller, analysis of more than 100,000 AI conversations a month showed that AI is especially helpful at the exploratory start of planning, when the trip is still taking shape, but travelers are not ready to make a decision from a handful of AI suggestions alone. When destinations and dates are still vague, they shape the trip through conversation; when it comes time to compare and commit, they return to the results pages and filters they already know. The conversational entry point and traditional search were being used side by side, but disconnected, inside the same traveler journey.
KAYAK concluded that this disconnect was a user experience problem, not a content problem. Bridging the gap between inspiration and booking required neither abandoning chat nor discarding search, but making the two work together on the same screen. Keller put it plainly in his interview with CX Today.
The right model is not chat instead of search, but chat working alongside search.
Given that so many AI travel tools have positioned themselves as replacements for the traditional search-and-filter experience, this amounts to a pointed dissent within the industry. Backed by its own usage data, KAYAK's position is that the replacement approach misreads where travelers actually need help.
People Do Not Book from a Handful of Suggestions: Designing for Trust
Trust is the wall every conversational interface in travel has to climb. Booking a flight or hotel involves real money and plans that cannot afford to fail. When a chatbot returns a few options with no broader context to cross-reference, choosing one becomes a leap of faith for the traveler.
Ask AI's answer was to ground the AI's suggestions in live, bookable search results rather than curated recommendations. Every time the user refines the brief in chat, the results list on the same screen updates instantly. Travelers keep the ease of conversation while comparing prices, timings and cancellation policies with the transparency they are used to.
Launching right as summer search demand ramps up was also a deliberate read on shifting traveler behavior. According to Keller, travelers now take longer to commit, track prices more closely and scrutinize cancellation policies earlier in the planning process. For iterative behavior that keeps adjusting conditions and reconsidering, a traditional UI that forces a fresh search each time carries too much friction. Ask AI is designed around that iteration itself.
An Evolution of Metasearch, or a Replacement?
Is Ask AI a one-off feature or the start of a structural shift? Keller's answer is both. It is a feature travelers can use today and a signal of where travel search is heading longer term. The experience will become more conversational and more personalized over time, but the company frames that as an evolution of metasearch rather than a replacement for the core tools travelers trust when it is time to book.
This careful positioning comes into sharper focus alongside parent company Booking Holdings' moves. KAYAK has taken a staged approach, testing cutting-edge features in its KAYAK.ai sandbox and folding mature ones into the main product. Unlike Expedia and Booking.com, which shipped apps inside ChatGPT, KAYAK embeds the conversational experience in its own site, keeping direct access to consumer behavioral data. Booking Holdings is also preparing Lola, a conversational AI venture led by KAYAK co-founders Steve Hafner and Paul English, stacking up bets on conversation as the entry point for travel transactions.
Competition among metasearch players is converging in the same direction. With Skyscanner beta-testing natural language destination discovery, an industry once defined as "the place to search and compare" is collectively moving toward handling everything from conversation to planning and booking. What makes KAYAK's announcement stand out is not feature flash but the explicit design principle derived from 100,000 monthly conversations.
For operators in travel and booking businesses, the implications are concrete. What actually runs beneath the conversation is live search, and whether your offering appears on the AI's suggestion surface depends on supplying inventory and prices in real time, in machine-readable form. The second lesson concerns the design of AI-assisted commerce itself: the behavioral finding that pairing chat with an existing comparison UI carries users to a decision better than trying to close everything inside chat applies well beyond travel, to any business handing the entry point of its transactions to AI.
Conclusion
KAYAK's Ask AI puts forward a design that is simple yet grounded in behavioral data: conversational AI and traditional search working together on a single screen. What 100,000 monthly conversations showed is that AI excels at the entry point of exploration, while decisions demand the transparency of comparison. Keller's principle — chat does not replace search, it works alongside it — offers a data-backed counterargument in an AI travel race where replacement narratives have run ahead. The shift toward conversation as the entry point for transactions will not stop, but whether that conversation earns trust depends on how accurately it is grounded in live inventory and prices behind the scenes. For operators who handle bookings and transactions, that is exactly where the preparation needs to happen.
