Why I'll Never Use AI to Plan a Trip

If you know me at all, you know I love to plan a trip. I enjoy finding new things and figuring out how to piece an intinerary together like a puzzle. I know a lot of people aren't like that and there's growing popularity of using AI to plan trips, which seems convenient, but can lead to a worse trip than one where you just show up and wing it. 

Now, even if I didn't love planning every aspect of my trip, I am more than a little skeptical about using AI for anything. First of all, AI is terrible for the environment. Each time you ask it to do anything for you, it creates a ton of heat, which takes the equivalent of six bottles of water (70+ oz) to cool down the server. At some point, if we continue to use it more and more, whole communities will be facing water shortages, and water becomes increasingly necessary with global warming making our summers hotter and hotter and drought conditions being much more common. 

Museum that was highly recommended, but closed on the day we arrived

If the environmental impacts don't concern you, and they absolutely should, then know that AI can never preplace real life research. There have been so many cases where people have relied solely on AI to plan their trip and getting there only to find out that places aren't open while they are there are permanently closed. I used a travel AI site months ago to see what it would give me as ideas on what to do at a destination and it had more exposition than actual suggestions, and what suggestions it did find me, which I tried multiple times, were the most basic activities that I could have easily Googled and also found more interesting options in not much more time. 

Tip: If you are using Chrome and also detest the AI search results on Google, you can download the extension "Hide Google AI Overview" and you'll only get regular results.

This travel writer put ChatGPT to the test with specific prompts, only to learn that the AI could not understand even the most basic things like maybe getting to the attraction before closing time or not planning something when she was supposed to be on her flight home. The stat that OpenAI can only achieve 10% success on a complex travel itinerary is not surprising (to me) and also means a lot of disappointment. 

With AI being so prevalent right now, there are a lot of companies using it as well. Those travel photos of your hotel or view from your window may be AI-generated. If you can't tell the shiny quality of the photos apart from real photos, that can be a real problem, especially when the hotel doesn't look anything like those images and you've paid a lot of hard-earned money to stay there. I always check to see actual customer photos and reviews for every place I stay. 

Not that pretty, but real hotel photo

You might even get fooled into thinking a place is cooler than it is, because so many people are also posting AI images of destinations on social media. It can be tough to tell if something is real or not, but the best way to do that is to Google image search it to see if there are any actual travel sites posting about it, not influencers. You can also tell by how the lines usually don't match up, parts of things are missing, and the pure absurdity of elements. 

I don't know about you, but I'm not a fan of having my time wasted, getting stranded, or only going to the most basic of tourist sites. If I have to spend a bunch of time trying to correct AI's mistakes on my trip, it's not worth it. I might as well have just spent that time before I go to make sure my trip is the best it can be while I'm on it. 

I don't think technology will ever replace real people, especially when I know that I may need to take a nap when I arrive and I'll want more than 10 minutes to enjoy an attraction. If I see some discrepancies, I can cross-reference sites or even call the company to confirm their hours or that they're even open at all. I use Google Maps and blog recommendations has a starting point for my trips.

Have you used AI to try to plan a trip? What was your result?