How we use artificial intelligence

1. What the AI does, and what it does not

Artificial intelligence writes the text of your itinerary: which stops to include, on which day and in what order, the descriptions, the tips, the notes on budget and packing. It is the creative part of the work — the part that saves you the hours you would spend fitting a day together.

It does not do all of it, and we want to say so. Places come from real databases (Wikidata and OpenStreetMap), the weather from Open-Meteo, the routes drawn on the map from OSRM: those it does not make up. Travel times, on the other hand, are proposed by the model and then re-checked against the real distance between one stop and the next: if they do not hold up, we correct them before you see them.

2. What it can get wrong

A language model can be wrong, and ours is no exception. The most frequent mistakes concern opening hours, prices, closing days and places that have changed their rules in the meantime. Less often, it may attribute to a place a detail that does not belong to it.

That is why every itinerary says that times and prices are estimates, and why we repeat it here: before you leave, and above all before you book, check on the venue's own website. The itinerary is an excellent starting point, not an official document.

3. If you find a mistake, tell us

We genuinely care. A precise report — which stop, which city, what did not add up — lets us improve the engine's instructions for everyone who comes after you. Use the site's contact form: we read everything.