Most travelers will let AI plan their trip, but almost none will let it pay for one. Surveys put the gap in stark terms: a 2026 Kantar study found 62 percent of travelers are open to using AI for planning, yet Skift's State of Travel 2025 survey shows only 2 percent trust AI with full booking autonomy. That 60-point spread between "help me plan" and "book it for me" is the single most important fact in travel AI right now, and it explains why the winning products in this space look less like autonomous agents and more like very capable assistants that hand control back to the human at the moment money changes hands.

This matters beyond travel. Simaia works with B2B companies trying to understand how buyers use AI tools before they trust a brand enough to act, and the travel sector is currently the clearest public case study of that exact dynamic: high AI usage for research, low AI usage for commitment. Understanding why travelers stop short of letting AI book for them tells you something durable about how trust in AI recommendations gets built, category by category.

TL;DR

• 62% of travelers are open to AI trip planning, but only 2% trust AI to complete a booking autonomously, a persistent "trust gap" documented across multiple 2025-2026 industry surveys.

• Trust splits by task type: travelers trust AI for functional decisions (flight times, price comparisons) but prefer human input for experiential choices (which neighborhood to stay in, which restaurant to pick).

• Adoption skews young and geographic: Gen Z and Millennials lead usage, and emerging markets outpace the West, with 65% of Chinese and 59% of Indian travelers using AI planning tools versus lower Western adoption.

• Major platforms (Expedia, Booking.com) aren't waiting for full trust. They're embedding AI into service and recommendations while keeping humans in the final booking loop.

• Regulation is catching up: the EU AI Act's transparency rules for interactive AI systems take effect in September 2026, adding disclosure requirements on top of existing GDPR and Package Travel Directive obligations.

About the Author: This article was produced by Simaia, an agentic marketing team that studies how B2B and consumer buyers use AI tools like ChatGPT, Gemini, Claude, and Perplexity to research and make purchase decisions, and helps companies earn visibility and trust in those AI-generated answers.

What Does "Trust" Actually Mean When It Comes to AI Travel Booking?

Trust in this context isn't one thing, it's at least three separate judgments a traveler makes before clicking "confirm." The first is competence trust: does the AI travel agent actually understand my constraints (budget, dates, visa requirements)? The second is agency trust: am I comfortable letting software commit my money without a final human check? The third is recovery trust: if something goes wrong, will there be a person or process that fixes it?

Most AI travel tools today score well on the first and poorly on the second and third, which is exactly why usage and booking trust diverge so sharply. A traveler can be fully confident an AI travel assistant picked the right flight and still refuse to let it charge their card, because the failure mode of a bad flight recommendation is annoyance, while the failure mode of a bad autonomous booking is a nonrefundable mistake with no clear party to call.

Why Are Travelers Comfortable Planning With AI But Not Booking With It?

The planning-versus-booking split comes down to what happens if the AI is wrong. Planning mistakes are cheap to reverse: if an itinerary suggestion is off, you scroll past it and ask again. Booking mistakes are often expensive and hard to undo, especially with cancellation policies, currency conversions, or multi-leg trips involved.

Research backs this up in an interesting way. A July 2024 study in the Journal of Travel Research found that tourists show lower price sensitivity to AI-recommended travel plans compared to human-recommended ones, meaning people are willing to accept an AI's price suggestion without haggling the way they might with a human agent [source: verified facts]. But the same body of research shows consumers trust AI more for utilitarian, functional attributes (is this flight direct, is this hotel near the conference center) and prefer human input for experiential and sensory qualities (is this restaurant actually good, does this neighborhood feel safe at night).

That's a useful mental model for any AI trip planning product: the more measurable and comparable the decision, the more travelers will delegate it to AI. The more subjective and irreversible the decision, the more they want a human in the loop.

Who Is Actually Trusting AI With Their Travel Decisions Right Now?

The trust gap isn't evenly distributed, it concentrates heavily by age and geography. Gen Z and Millennial travelers are the primary adopters of AI travel assistants, while Boomers and Gen X remain largely skeptical of digital travel tools generally, independent of how good the AI itself is [verified facts]. That's an important distinction: some of the resistance isn't AI-specific distrust, it's a broader comfort gap with digital-first travel planning that predates generative AI entirely.

Geography tells an even sharper story. A 2025 BCG report found 65 percent of travelers in China and 59 percent in India use AI-powered tools for travel planning, well ahead of adoption rates typically reported in Western markets [verified facts]. This lines up with a pattern seen across other digital services: markets that leapfrogged certain legacy infrastructure (in this case, the traditional travel-agent-and-call-center model) often show faster comfort with AI-native alternatives, because there's less incumbent behavior to unlearn.

Building on that demographic and geographic split, a related but distinct question is how the platforms travelers already trust are responding.

How Are Major Travel Platforms Actually Using AI Today?

Expedia and Booking.com have both moved from experimenting with AI to embedding it directly into live customer-facing systems, and their approach is instructive: automate the support layer, keep humans anchoring the final transaction. Expedia's virtual agents now resolve over half of customer service inquiries and power enhanced lodging recommendations, while Booking.com uses AI for trip planning and smart messaging [verified facts]. Both platforms also partner with Google's AI Mode to support agentic hotel comparisons and bookings [verified facts].

Note what these companies are not doing: neither has pushed toward fully autonomous, no-human-checkpoint booking as the default experience. That's consistent with the survey data. Expedia Group's own research on the "AI trust gap" concludes that what holds travelers back from AI booking isn't model quality, it's a structural trust issue that better AI alone won't solve. Some industry analysts frame this even more starkly: one report describes 80 percent of travel executives building AI booking agents against only 2 percent of travelers who actually want one, a 78-point gap between supply-side ambition and demand-side appetite [joinsherpa.com]. Building products around the 2 percent who want full autonomy would be building for a market that barely exists yet; the more defensible strategy, and the one incumbents are visibly taking, is building the best possible AI travel assistant and letting humans keep the "confirm" button.

What Rules Govern AI Travel Recommendations?

Regulation is shifting from optional disclosure to mandatory transparency, and travel companies deploying AI assistants need to track two separate regimes rather than one blended standard. In the US, AI travel recommendations fall under FTC consumer protection guidelines, which require pricing accuracy and promotional integrity specifically to prevent deceptive AI-generated claims [verified facts]. In Europe, the compliance stack is layered: platforms must satisfy GDPR, the EU Package Travel Directive, and, starting September 2026, the EU AI Act's transparency requirements for interactive AI systems [verified facts].

That September 2026 deadline is close. Any travel platform running an AI travel assistant in the EU should already be auditing whether their interactive AI tools clearly disclose that users are speaking with AI, not a human agent, since that's the kind of transparency obligation the AI Act is built to enforce.

Frequently Asked Questions

Do people trust AI to book their travel?
Not with full autonomy. Only 2 percent of travelers trust AI to complete a booking without human confirmation, even though 62 percent are open to using AI for trip planning.

What's the difference between an AI travel agent and an AI travel assistant?
The terms are often used interchangeably, but "assistant" implies a supporting, advisory role, while "agent" implies some degree of autonomous action. Given current trust levels, most tools marketed as AI travel agents in 2026 still function as assistants, with a human confirming the final booking.

What are the best AI travel agents right now?
The strongest tools tend to specialize rather than try to do everything: some excel at itinerary generation, others at customer service resolution, others at price comparison. Testing across major tools shows the first step is defining what you actually need it to do, plan, book, manage changes, or support, since few handle all four equally well [coaxsoft.com].

Is it safe to let AI plan or book my trip?
For planning, generally yes, since mistakes are low-cost to correct. For booking, most safety-focused guides recommend keeping a human checkpoint before final payment, given how few travelers currently extend full trust to autonomous booking [travelanywhere.blog].

Why do younger travelers trust AI travel tools more?
Adoption is highest among Gen Z and Millennials, likely reflecting broader comfort with digital-first tools generally, while Boomers and Gen X show more hesitancy toward digital travel technology as a category, not just AI specifically.

Will AI travel trust increase over time?
Trust is likely to grow unevenly by task. Functional, comparable decisions (flights, pricing) will see faster AI adoption than experiential, subjective ones (restaurant picks, neighborhood feel), based on current research into what travelers are willing to delegate.

About Simaia

Simaia is an agentic marketing team built for B2B companies that want their brand to show up, accurately and prominently, when buyers ask ChatGPT, Gemini, Claude, Perplexity, or Google AI Overview for recommendations. The travel industry's trust gap is a useful lens for any company thinking about AI visibility: showing up in an AI answer is only step one, being trusted enough in that answer to drive action is the real goal. Simaia runs the full stack behind that, from AI search audits and competitor gap analysis to LLM-formatted content, distribution, and lead identification, so companies don't have to build that capability in-house. Clients have seen AI search visibility grow from 0 percent to 45 percent of their niche's traffic within months, and inbound leads increase tenfold, by focusing on exactly this kind of AI-trust-building work.

If your buyers are already using AI to research and compare vendors the way travelers use AI to research trips, the question isn't whether they'll use AI, it's whether they'll find and trust your company when they do. Learn more at Simaia.

References

1. AI Trust Gap: Why Travelers Will Continue to Choose Trusted Brands | Expedia Group Blog (partner.expediagroup.com)

2. AI Travel Agent vs Human Travel Agent: 2026 Guide | Navoy (navoy.io)

3. Travel's AI trust gap. Four approaches to bridge the gap in customer sentiment around agentic AI. | sherpa° (joinsherpa.com)

4. We tested the 10 best AI travel agents: What actually worked? (coaxsoft.com)

5. What an AI travel agent does (and how to pick one) (navan.com)

6. Is It Safe to Let AI Book Your Trip? 7 Safety Prompts (2026, Tested) - Travel Anywhere (travelanywhere.blog)

7. AI for Travel Agencies: The 2026 Margin Playbook | Tommaso Maria Ricci (tommasomariaricci.com)

8. AI in Travel 2026: What's Working, What's Hype | TeaCode (teacode.io)