Travel AI is fragmenting
Big tech firms are building competing travel planners while industry leaders say GenAI is strongest at discovery, not bookings, creating a split between discovery agents and booking/fulfilment systems. Skift flags fragmented travel planners from Amazon, Meta and Google, and MakeMyTrip’s CTO notes chatbots excel at discovery while OTAs retain booking and post‑sales dominance. That division creates distribution and integration challenges for travel platforms trying to be both discoverable and authoritative on transactions. (x.com) (x.com)
Ask three different artificial intelligence assistants for a trip to Tokyo, and you may get three different “best” neighborhoods, three different flight options, and no clear place to fix a canceled booking. That is the split opening up in travel right now. (skift.com) Skift reported on April 10 that travel is not dealing with one new gatekeeper but several, with Amazon, Meta, and Google each building their own travel planning paths. Each system works differently, so a hotel or airline may need to show up in multiple artificial intelligence ecosystems instead of one search box. (skift.com) Google has already pushed this furthest in public. In a November 17, 2025 post, Google said its Search “Canvas” tool can build itineraries with flight and hotel data, Maps reviews, and web results, and its Flight Deals feature is rolling out in more than 200 countries and territories in over 60 languages. (blog.google) Amazon is taking a different route through Alexa Plus, the upgraded voice assistant it began showing off in early 2026. Skift reported on April 1 that Amazon sees travel booking as a next step after teaching Alexa Plus to complete simpler commerce tasks like ordering pizza. (skift.com) Meta’s angle is not a booking engine first. Skift reported on April 8 that Meta’s new artificial intelligence model could reshape travel discovery by combining what people see, share, and ask across Meta’s apps with real-time recommendations. (skift.com) That sounds powerful until the traveler needs something boring and difficult, like changing one leg of a multi-city ticket after a delay. Sanjay Mohan, the group chief technology officer of MakeMyTrip, said on April 10 that generic chat systems are good at discovery, while online travel agencies still own booking, transactions, and post-sales service. (bestmediainfo.com) In plain terms, discovery is the part where a machine helps you dream, compare, and narrow down options. Fulfilment is the part where someone has to hold inventory, collect payment, issue the ticket, apply refund rules, and answer when the airline changes the schedule. (bestmediainfo.com) Travel companies have spent two decades building those fulfilment pipes inside online travel agencies, airline sites, hotel systems, and call centers. Artificial intelligence assistants are now trying to sit one layer above that plumbing, which is why the industry suddenly has a discovery layer and a transaction layer pulling apart. (skift.com) That creates a distribution problem for suppliers. A hotel group now has to ask whether it should optimize for Google’s itinerary builder, Amazon’s voice flow, Meta’s recommendation graph, or its own direct booking channel, because the answer may be all four. (skift.com) It also creates a trust problem for travelers. If one assistant inspires the trip, another surfaces the fare, and a third party handles the ticket change, the smooth chat experience can break the moment money moves or something goes wrong. (bestmediainfo.com) That is why the fight is shifting from “who has the smartest chatbot” to “who controls the handoff.” The winner in travel may not be the assistant that writes the prettiest seven-day itinerary, but the company that can carry a traveler from “Where should I go in October?” to “Your refund has been processed” without losing the thread. (skift.com)
