How to use AI to guide your holiday plans – by a tourism expert

If you ask an AI service like ChatGPT or Google Gemini to recommend a destination for your next summer holiday, it will happily provide you with a list of attractive destinations. But many of them will be very familiar.

Paris, Venice, Santorini and Barcelona are all likely to feature, because the AI algorithm is nudging you towards the same old places. The illusion of personalised advice is what makes people less likely to question it – and why AI risks intensifying overtourism^[1].

And the use of AI for holiday inspiration is growing fast. A recent survey^[2] found it has doubled in the past year, with uptake strongest among younger travellers. Nearly one in five Britons aged 25–34 now turn to AI tools to plan their trips.

In my own research^[3], I analysed ChatGPT’s travel recommendations and found that it gravitates towards the most visited destinations by default. Lesser-known or more sustainable locations only tend to appear when travellers explicitly ask for them.

This could easily exacerbate the overtourism which is already testing the limits^[4] of many residents in highly visited places. In Mallorca, locals are demanding limits on flights and holiday rentals, while Venice introduced a day-tripper fee^[5] in an attempt to manage visitor pressure.

AI will quickly add to that pressure if millions of holiday makers make plans using the same online filters and tips. These algorithms are trained on what’s most visible online – reviews, blogs and social media hashtags – so quickly focus on what’s already popular.

And if travellers simply accept the defaults, the result will be more of the same, and more strain on places already under pressure.

But consumers aren’t entirely powerless. With a bit more intent, AI research can yield different and fascinating destinations.

My research suggests that discerning travellers need to start by asking better and more searching questions. Generic prompts such as “the best beaches in Europe” or “beautiful city” lead straight to the same results.

Instead, try something like: “Which towns are reachable by train but overlooked in most guides?” Or maybe: “Where can I go in July that’s not a major tourist hotspot?”

Push the system, ask follow-up questions and scroll past the first few results. That’s where the surprises often lie.

You could also change your timings. AI tends to focus on peak season because that’s when the most online reviews are posted and the most travel content is published.

Asking about off-peak months is a simple way to beat this built-in bias, so perhaps specify the Italian lakes in October or the Greek islands in May.

Or ask AI to dig a little deeper for its source material. AI draws heavily on English-language content, which favours international hot spots, but is also capable of finding independent travel blogs or local tourism cooperatives.

Type in something like “Spanish-language blogs about Asturias” or “community-run agritourism in Slovenia” and you could unearth something rewarding and off the beaten track. This is the kind of thing that can really unearth the vast potential benefits of AI and its capabilities.

It could also easily help you to compare the costs and timings of various travel options, and assess the carbon footprint of your journey. It just requires a little bit of digging to get past the surface layer.

After all, these systems are designed to serve up the most obvious and well-documented suggestions, not what’s diverse or sustainable. (Although the same technology could just as easily be coded slightly differently to show rail travel before air for example, or to prioritise locally run independent businesses.)

So while the convenience of AI is seductive, it can also be predictable. If your holiday plans could be copy-pasted from Instagram, any sense of adventure can easily get left behind.

Consider using AI as a starting point, not the final word. Guidebooks, local media and conversations with residents restore the unpredictability that makes travel memorable.

By asking sharper questions, shifting their timing, checking footprints and seeking local voices, travellers can use AI as a tool for discovery rather than congestion. Every prompt is a signal to the system about what matters.

The next time you ask ChatGPT where to go, make it work a bit harder. Test it, argue with it and use its extraordinary capabilities to find somewhere new – or settle for the same crowded itinerary as everyone else.

References

1. ^{^} intensifying overtourism (theconversation.com)
2. ^{^} recent survey (www.abta.com)
3. ^{^} my own research (www.tandfonline.com)
4. ^{^} testing the limits (theconversation.com)
5. ^{^} a day-tripper fee (theconversation.com)
6. ^{^} organtigiulia/Shutterstock (www.shutterstock.com)