Representative ImageArtificial intelligence (AI) tools used for holiday planning may be biased towards well-known destinations, according to a study by SALT.agency. The research analysed Google's AI responses to approximately 85,000 travel-related queries, revealing a tendency to recommend familiar locations such as Canada, Mexico, the USA, India, and Australia. This bias could lead to overcrowding and higher costs at these destinations, whilst travellers may miss out on discovering hidden gems.
The study highlights that AI recommendations are not evenly distributed, creating a "winner takes most" scenario. For instance, Canada was suggested in nearly 10% of the 65,000 general travel planning responses analysed. Dan Taylor, Partner and Head of Innovation at SALT.agency, noted, "AI tools have a habit of recommending the same places over and over, being specifically biased to ‘well-known’, ‘well-documented’, and ‘easy to justify’ destinations."
For luxury travel, the top destinations identified were India, Italy, the Maldives, Singapore, and Australia. Meanwhile, solo travellers were most commonly directed to Japan, Thailand, Portugal, Costa Rica, and Iceland. Taylor warns that AI's tendency to streamline travel planning could "shrink the world" by narrowing options and discouraging exploration. "AI is excellent at producing a polished itinerary, but that itinerary is often built from the same ingredients," he added.
The findings suggest that whilst AI can simplify travel planning, it may also limit the sense of discovery that many travellers seek. For more insights, visit SALT.agency's blog.
```
