There is a version of travel that the industry loves to talk about right now. You open an app, type a few sentences about yourself, and within seconds an AI has mapped out your entire trip. Hotels, experiences, restaurant bookings, transfer times. It looks clean, confident, and effortless.
But most of the trips people actually care about are not solo ventures planned on a weeknight. They are the family holiday saved for all year. The anniversary trip promised for longer than either of you would admit. The first time abroad with the kids, the one they will describe to their own children one day. Travel at its most meaningful is rarely just about you. It is about the people sitting next to you on the plane, and whether the thing you planned so carefully is going to hold.
That is why what I am about to describe is not a product complaint. It is something the industry needs to honestly reckon with.
I have spent the better part of a decade working in and around travel technology, first at Dubai Tourism, then building platforms that sit at the intersection of content, booking, and personalisation. And the one thing consistently overlooked in every conversation about AI and travel is the moment things go wrong. Not the itinerary. The recovery.
Where AI Falls Short: The Moment of Disruption
The scale of what is at stake in 2026 is significant. Global tourism is projected to reach 1.58 billion international arrivals this year, a new all-time record. Dubai welcomed nearly 20 million visitors in 2025, its third consecutive record year, and has declared 2026 the Year of the Family. Japan drew over 42 million international visitors in 2025, a 16 per cent rise year on year. These are not cautious numbers. People are travelling, and they are travelling with the people they love most.
And increasingly, they are planning those trips with AI.
A McKinsey and Skift joint study found that extensive AI use for trip planning more than doubled in a single year, rising 124 per cent between 2024 and 2025. People have taken to these tools fast and with real enthusiasm.
What has not kept pace is accountability for when those tools get things wrong. And they do. Squaremouth's research found that while 47 per cent of travellers have used AI to build itineraries, one in three reported receiving false or misleading information in the process. These are families who arrived at attractions that were closed. Couples who showed up at restaurants that had shut months earlier. Judy Gauthier of Go City described being guided by an AI planner to a trail in the Smoky Mountains, only to find through a Facebook post that it had been closed for 18 months. The AI had no way of knowing. It said nothing about it.
With limited access to live calendars, updated opening hours, or real-time supplier data, these AI systems give answers based on low quality online data and patterns. And because these LLMs communicate in an authoritative, unhurried tone, travellers extend them a degree of trust they may not deserve yet.
The Trust Gap: Why Travellers Still Hesitate
That trust may already be starting to wear thin. KPMG's 2025 global study, which surveyed over 48,000 people across 47 countries, found that although 66 per cent of people now use AI regularly, less than half, just 46 per cent, are willing to trust it. More telling still, people have become less trusting and more worried about AI as adoption has gone up. That is the opposite of what you would normally expect. It tells you something specific about where the real problem lies.
McKinsey and Skift's research frames it differently but arrives at the same place. More than 90 per cent of travellers trust the information AI gives them for inspiration and planning. Yet only 2 per cent are currently willing to let an AI tool make and modify bookings on their behalf without a human involved. That gap between trusting the plan and trusting the recovery is not random. It is travellers telling us, through their behaviour, exactly where their confidence runs out.
I have built enough in this space to believe in what AI can do for travel. When it works, it genuinely changes the quality of the planning experience. The ability to surface a personalised itinerary for a first-time traveller to Japan, or to help someone piece together a complicated multi-city trip, is real value. I have seen it.
But there is a meaningful difference between a tool that helps you plan and a platform that takes responsibility for your journey. What needs to be built next is not a smarter itinerary engine. It is the infrastructure of accountability. What happens when a booking fails? Who reaches the traveller? How quickly? What is the path to resolution when a family of four lands in Tokyo and the hotel has no record of their reservation? These are not glamorous product questions. They do not get featured in a launch announcement. But they are the questions that will determine whether AI-led travel earns lasting trust or whether people go back to booking with a human, because at least a human picks up the phone and aims to resolve a complicated situation.
In software, there is a concept called a self-healing system. When something breaks, the system identifies it and corrects it without waiting for a person to step in. Travel AI has no equivalent of this. It plans forward beautifully. Disruption is an oversight. McKinsey has pointed out that the technology exists to do better, to reach travellers in real time when operations go wrong, to offer solutions rather than silence. The vision is there. The build is not.
The question the industry needs to start answering is not how do we get the traveller to the booking. It is what are we prepared to do when something goes wrong after they get there.
Until that question has a real answer, accountability remains the missing layer. And the people bearing the cost of that are not the ones building the AI itinerary apps. They are the couple standing in an unfamiliar city at nine at night, the kids exhausted beside them, staring at a screen, wondering why the AI that planned everything so confidently has nothing to say now.
That is the problem worth solving.
FAQs
1. What is travel AI?
Travel AI refers to artificial intelligence tools that help users plan trips, generate itineraries, and recommend hotels, activities, and routes.
2. How is AI used in travel planning?
AI is used to create personalised itineraries, suggest destinations, optimise routes, and provide recommendations based on user preferences and data.
3. What are the limitations of AI in travel?
AI often relies on outdated or incomplete data, lacks real-time updates, and cannot always handle disruptions like cancellations or booking errors effectively.
4. Why is accountability important in travel AI?
When travel plans fail—such as incorrect bookings or closures—travellers need immediate support and solutions. Without accountability, trust in AI tools declines.
5. Can AI handle travel disruptions?
Currently, most AI tools are better at planning than resolving issues. They lack the infrastructure to provide real-time assistance or recovery solutions.
Ashish Sidhra is the Co-Founder of Alike.io, a global, AI-driven travel platform.
