Amanda Smith is a freelance journalist and writer. She reports on culture, society, human interest and technology. Her stories hold a mirror to society, reflecting both its malaise and its beauty. Amanda's work has been published in National Geographic, The Guardian, Business Insider, Vice, News Corp, Singapore Airlines, Travel + Leisure, and Food & Wine. Amanda is an Australian living in the cultural center of gravity that is New York City.
Vacation planning has come a long way since travel agents were commonly used. These days, we build our bucket lists from Instagram hotspots, TikTok trends and YouTube influencers.
But now, with artificial intelligence tools, travel planning is becoming much more personalized. Travelers are environmentally aware and seeking a deeper meaning from their adventures. AI can be a helpful travel assistant, guiding you in your choice of destination based on what you're looking to experience.
But does a journey start with a feeling, not a destination? I'm using The Feelings Engine to find out.
The Feelings Engine is the AI-enhanced tool by award-winning travel experience creator Black Tomato. It launched in February and is free to use.
You can talk to the chatbot about how you feel and what you want to experience. For example, "I want to feel the thrill of tracking wildlife" or "I want to feel like I'm stepping into a storybook." You can follow up an initial recommendation by asking questions about the best time to go, accommodation ideas, the type of cuisine and, what makes it family-friendly or perfect for solo travelers.
With a world of possibilities on where you can go, leaning on AI created by leading travel experts is a fun place to start your research.
The Feelings Engine's model is trained on two decades of its own proprietary data, including hundreds of its website itineraries and their underlying classifications to curate tailored travel experiences based on your emotions. For example, the core feelings each trip invokes, the category of experience and the types of travelers they're suitable for.
It is run through the Double Retrieval-Augmented Generation system -- two systems that are designed to improve accuracy and reliability. The initial RAG queries the data fed it with all the touchpoints above, while the second RAG (GroqCloud) assesses the accuracy of the responses.
If you have no idea where you want to go, you can start your Feelings Engine search broadly. For example, you could start with the following prompt: "I want to feel free."
I started with a broad prompt to see what the platform could generate: "I want to feel connected." It suggested a 10-day trip through Nepal and Bhutan -- check out all the details it included below:
Not bad. Nepal has been on my list for ages, especially the neighboring Tibetan region. At the bottom of the response, there are "Explore Trip" and "Get in Touch" buttons. To test out its chat capabilities, I replied that I'm in New York and asked if there's a similar experience that's closer to me.
And it did well -- suggesting Utah, which is also on my list:
Wow. A quick Google Image search of Amangiri, the accommodation recommendation, and I'm sold.
A hotel built around a canyon? Yes, please. The only problem is that it's $5,000 per night, and the Kardashians blew it up, so it's beyond my budget. I told that to the chatbot.
At this point, I'd probably talk to human travel experts about how I can still visit Utah at a more affordable price point. Spending $5,000 all up for a domestic trip is OK, but not $5,000 per night.
But Black Tomato said that "as a luxury travel company," its specialty is in curating high-end, personalized experiences for its clients. It suggested connecting me with a human travel expert and asked if I would prefer a different destination suggestion.
However, since I was happy with the location the model suggested, I felt we were off to a good start.
A new browser opened with a contact form when I clicked Get in Touch. I could see the minimum per person spend for any vacation was £4,000, which is equivalent to about $5,000 -- so keep in mind these will all definitely be premium experience suggestions.
I returned to the original browser and started a second search. This time, I had an idea of the type of trip I wanted, so I could give it more context. My wife and I never had our honeymoon, as we got married during COVID. We're also about to start IVF, so we want to go on a trip once we're pregnant to celebrate. A honeymoon/babymoon trip.
I fed the AI tool that info, and asked for "somewhere safe, close and relaxing for about one week."
It suggested Iceland, which is again a great option. But since I've been there before, I asked for some other destinations.
While it didn't provide a list like I requested, it did suggest Vienna in Austria, Croatia, Spain and France. I replied that I feel an international trip is a bit ambitious and to provide options closer to home.
It was then that I realized it didn't connect that "home" is New York, not the UK (where the company is based), so I had to specify.
Then, it suggested a staycation -- maybe too close to home now -- or venturing further out to upstate New York, the Berkshires or along the New England coast.
We want to stay in North America but also want it to feel like we're traveling. Somewhere like Québec City. Let's see if Feelings Engine suggests it. I added that we wanted to fly there, and it should be somewhere romantic.
It suggested Napa or Sonoma in Northern California, which is a great idea, but we've recently done a road trip up the Pacific Northwest Highway. I replied, saying we only wanted to fly for 1 to 2 hours. While it suggested Palm Springs -- a 6-hour flight -- it did list Montreal down the bottom, so we're getting close.
I followed up by asking if Québec City is a good option.
While a cross-country Canadian adventure was too much, a week exploring Montreal and Québec City sounds dreamy.
I told this to the chatbot, and we ended our conversation here.
I preferred using the tool when I didn't have a destination in mind because it felt more exploratory. Now Utah is top on my local bucket list, which is something I couldn't say before.