How AI Is Powering Travel Right Now (and What It Means for the Future)
Travel planning is in the middle of a major transformation, and it’s happening fast. For decades, travelers relied on guidebooks, blog posts, and traditional search engines to map out trips. Now, conversational AI is increasingly becoming the first place people go for recommendations, itinerary-building, hotel research, and even booking decisions.
From ChatGPT-style trip planning to AI-driven hotel discovery and next-gen recommendation engines, artificial intelligence is quickly becoming the new travel “front desk.” But as the tools get more intelligent and more persuasive, the industry is facing serious questions around accuracy, trust, privacy, and whether AI will worsen overtourism by sending everyone to the same places.
To understand where this is going and how travel brands are adapting, we spoke with Thomas Hertkorn of a&o Hostels about AI-powered travel, who offered an on-the-ground view of how AI is reshaping traveler behavior, hospitality marketing, and the future of human expertise in tourism.
AI-Powered Travel and Search Tools Are Becoming the New Default
AI has changed the psychology of travel planning. Instead of typing “best hotels in Berlin” and clicking through 15 links, travelers can now ask a single question like: “Where should I stay in Berlin if I’m traveling solo, want a social vibe, but need quiet sleep?” and get a tailored response instantly.
This is why AI is no longer just “tech.” It’s a new travel distribution channel, and it’s influencing everything from what travelers expect to how they review experiences after the trip.
Thomas Hertkorn of a&o Hostels says the shift is already measurable: “Currently, around 4% of our direct bookings are generated via ChatGPT.” He also notes that AI is improving alignment between expectations and reality, which is critical for budget-focused hospitality: “With AI, their decision is more informed… Such details educate guests and lead to better reviews because they get what they expect.”
In other words, AI isn’t just inspiring travel; it’s pre-setting the guest’s mindset before arrival.
The Biggest Risk: AI Hallucinations and Travel Misinformation
AI can be brilliant at summarizing, comparing, and suggesting. But it can also be dangerously confident when it’s wrong, especially when there isn’t enough high-quality information available about a destination or supplier.
That problem hits small and independent travel operators hardest.
Hertkorn points to a growing issue: “The biggest challenge is the false information flying into AI, especially when there is no trustworthy source for it.” He explains that when chatbots can’t find enough information, they sometimes fabricate: “If an AI chatbot doesn’t find information, or enough information, it will start to hallucinate and make things up, which leads to false information.”
And this isn’t hypothetical. He shares an example from hospitality that shows how bizarre it can get: “At a&o Hostels, for example, we have seen five-star reviews for our whirlpools, despite not having whirlpools at any of our properties across Europe.”
This is the core travel AI tension: personalization is amazing, but verification is everything. In a world where travelers trust a single chatbot answer more than five tabs of research, accuracy becomes a brand survival issue.
AI Bias in Travel: The Hard Truth Nobody Wants to Say Out Loud
One of the biggest debates in AI travel is algorithmic bias: who gets recommended, who gets ignored, and which destinations are framed as “safe,” “cool,” or “worth it.”
But Hertkorn argues bias isn’t something you can fully delete, because it originates from the world AI learns from: “When we discuss ‘bias’ in AI, we are really talking about society. AI learns from humans, and everything we create that finds its way to AI becomes a learning tool.”
That has real consequences in travel discovery, especially for vulnerable traveler groups. He gives an example tied to solo female travel: “There is a lot of bias for female travelers when, for instance, a past incident results in AI not recommending a city as a travel destination even though it may have been a one-time thing.”
His takeaway is blunt: “One can’t change the bias in AI; one can only adjust it.”
For travel brands, this means the AI era is not only about SEO or content strategy. It’s also about ethics: what data is being used, how recommendations are formed, and whether the system is reinforcing unfair narratives about places and people.
Privacy Is Becoming a Competitive Advantage in AI Travel
As AI gets integrated into travel discovery, hotels and travel operators are facing a new kind of operational risk: feeding sensitive information into AI tools.
Hertkorn says a&o Hostels treats this seriously, especially when using tools like ChatGPT: “For ChatGPT, for instance, we have standard agreements that prohibit the use of data that we share.” He adds: “[We have] an internal guideline that covers non-shareable information, such as the names of clients and their financial and booking information.”
He also points out that privacy awareness is especially advanced in Europe: “The things we’re doing are widespread in Germany and Europe, because we are educated on working closely on the guidelines.”
The travel industry’s future trust problem won’t just be about scams or fake reviews. It’ll also be about whether travelers believe their data is safe in an AI-driven booking world.
Can AI Reduce Overtourism, or Will It Make It Worse?
A major criticism of AI travel tools is that they can become popularity machines. If a model is trained on reviews, social media posts, and widely published itineraries, it can keep recommending the same famous spots, concentrating crowds and making overtourism worse.
Hertkorn believes AI can do the opposite, but only if the prompts are designed thoughtfully. He explains that AI can be guided toward lesser-known recommendations by widening context: “We respond by asking them: ‘What is your moment of travel? What is the purpose of your travel? What is your state of mind?’”
Then, once those details are included, it changes the AI output: “Through their responses, we can adjust the prompt we send to ChatGPT and get more and lesser-known recommendations.”
His point is important: AI doesn’t automatically solve overtourism. Humans have to use it in a way that encourages dispersion, not concentration.
How to Get Better AI Travel Results: Stop Being Vague
The biggest mistake travelers make with AI is treating it like Google.
AI works best when it understands intent, not just location. Hertkorn says it plainly: “For travel, AI is intention-based and focused on what one wants to experience locally.” His advice: “If you’re looking for local dining, or the best party locations for a specific dance style, or are traveling with kids… the key to getting the best results is to give AI as much information as possible.”
In practice, that means travelers should write prompts more like a short brief than a keyword search.
Instead of:
- “Where should I go in Spain?”
Try:
- “I’m traveling solo for 7 days in Spain in late October. I want smaller cities with local food culture, Moorish history, and day hikes. I prefer boutique hotels, no nightlife, and I’m burned out from work.”
That’s how you get surprising results.
“Long-Tail Keywords” Are Dead in AI Travel
For years, SEO strategy taught travel brands to chase long-tail keywords like “best hidden beaches in Mallorca for couples.” But AI doesn’t work like that.
Hertkorn argues the entire concept is outdated: “By the way, ‘long-tail’ is a term from the past days of SEO. There is no long-tail in AI, because every prompt is unique.”
Instead of keywords, AI runs on context. And context creates clusters of intent. He adds, “It is more about finding the context and then building clusters, which AI is doing.”
This is a massive shift for travel marketing: content still matters, but it needs to serve AI discovery, not just page-one rankings.
Hyper-Personalized Travel: Moving Beyond Demographics
Traditional travel personalization has often been shallow: age group, budget, couple vs. family. But AI can personalize based on behavior, travel style, and sequencing.
Hertkorn says a&o Hostels is already moving in this direction: “We are moving away from travelers’ budgets toward traveler behavior.” He gives examples of behavioral signals: “How long does one want to stay? What is their travel time period? Are they thinking of combining several stays with us… or will they be moving around with us via rail pass.”
This is where AI travel is heading: not “people like you booked this,” but “people with your rhythm, mood, and movement patterns enjoyed this.”
AI Planning Is Fast, But Trust Still Wins
AI-powered travel can create a multi-city itinerary in seconds, but speed alone doesn’t make travel meaningful. Travelers still need emotional connection and confidence, especially for expensive trips.
Hertkorn says it directly: “Speed is great, but trust matters just as much.”
He also ties trust to transparency, including values-driven storytelling from brands: “From 2026 onwards, we will start sharing these stories more openly… We believe this transparency strengthens emotional connection and long-term trust.”
At the same time, he acknowledges that travelers may feel overwhelmed by the sheer number of competing AI systems: “The problem stems from there being so many AI providers at the moment: Google, Gemini, Booking, Trip.com, Expedia, ChatGPT… and more.” He predicts consolidation: “At some point, the market will develop a bigger, one-fits-all AI solution.”
Will AI Replace Travel Agents? Yes, But the Story Doesn’t End There
The hardest question in travel AI is the human one: what happens to travel agents, concierges, and local experts?
Hertkorn believes many traditional roles will be displaced: “Unfortunately, AI will replace a lot of human travel agents.” But he also sees a new evolution where the job becomes more creative and influence-based: “We’ll see movement away from simple, sanitized jobs to more creative jobs.”
And he makes a fascinating prediction about where travel agents may land: “I think in the end result will be classic travel agents becoming travel influencers because they have more knowledge and networks than any other travel influencers out there.”
The human role doesn’t disappear. It transforms into verification, taste-making, crisis response, and storytelling.
The Future of AI Travel: Your Phone Becomes Your Real Assistant
AI travel isn’t just about booking. It’s about the entire travel life cycle: dreaming, planning, purchasing, navigating, revising, and remembering.
Hertkorn paints the direction clearly: “We are heading to a world where a mobile phone will be one’s personalized assistant – and a real assistant, at that.” But he also warns what must keep pace: “Provided that transparency, data ethics and user trust evolve at the same pace as the technology itself.”
And perhaps the most human part of all: even the experts are trying to keep up. “Personally, I’m often overwhelmed by the sheer speed at which AI is evolving.” Still, he believes we’re moving toward a more stable phase: “I’m convinced the Gartner Hype Cycle also applies to AI, and that we are currently moving from peak expectations toward the plateau of productivity.”
Read more travel advice here!
