The trains were booked, the operator feels that his work has been done. But I want somewhere to live, I remind it – can it book a hotel? It asks for more details and I am objectively vague, refer to it that it should be comfortable and easily located. Comparing the hotel is probably my lowest favorite aspect of travel plan, so I am happy to quit scrolling through booking.com. I prevent myself from jumping when I see it determines the wrong dates, but it corrects itself. It spends some time in survey of an IBIS listing, but selects a three -star hotel called Martin Bruge, which I have noted that users have evaluated as an excellent location.
What is left now is a journey program. Here, the operator starts losing steam. It offers a full one -day program that appears to be primarily cribbed from a vegetarian travel blog. 2 days, it suggests that I “visit any remaining attraction or museums.” Wow, thank you for the tip.
The journey day comes, and, as I pull myself out of bed at 4:30 am, I remember why I usually avoid the early departure. Nevertheless, I go to Brussels without any issue. My ticket allows for further travel, but I realize that I don’t know where I am going. I fire the operator on my phone and ask from which platform the next brugs-bound train departs. It discovers the Belgium Railway time -to -Carini. After minutes, it is still searching. I see and see details on the station display. I reach the platform before finding out the operator.
Bruges are delightful. Given the lack of the operator’s lack of traveling, I bran out. Such research work is perfect for a large language model, I realize – it does not require agent capabilities. Chatgpt, Openai Sibling of the operator, gives me a lot of perfectly plan, not only to eat, but what to order to order activities by hours with suggestions to order activities (de Halle Mann Brevari in Flemish Stu). I also try the cloud of Google’s Gemini and Anthropic, and their plans are the same: run on the market square; See Belfree Tower; Visit the bourgeal of the holy blood. Bruges are a small town, and I cannot help, but wonder whether it is only the standard tourist route, or if the AI models are all getting their information from the same source.
Various travel-specific AI equipment are trying to break through this generality. I briefly try Mindtrip, which offers a map with a written journey program, offering personalizing recommendations based on a quiz, and includes collaborative characteristics for shared visits. CEO Andy Moss says it expands on broad LLM abilities by taking advantage of a travel-specific “knowledge base” that contains things such as weather data and real-time availability.
Courtesy of Victoria Turk
