Five models, one score sheet
MEININGER Hotels put a holiday question to five AI systems: ChatGPT, Claude AI, Gemini, Microsoft Copilot and Perplexity AI. The company took the 100 German cities and the 100 European regions with the highest annual overnight-stay counts and attributes the basis to Regionalstatistik and Eurostat for 2024. Each model scored every destination from 0 to 100, and MEININGER averaged the five scores into one figure per place. The prompts asked the models to avoid increments of five and ten. A second prompt asked for the one annual event a traveller has to see in each destination, and a third scored those events on the same scale.
Heidelberg and Rome lead
Heidelberg finished first in Germany on 94.4 points. Rome finished first in Europe on the same 94.4. Dresden and Freiburg im Breisgau tied for second place in Germany on 92.6. Munich scored 91.8 and Bamberg 91.2. The rest of the German top ten runs Lubeck 90.8, Berlin 89.4, Regensburg 88.4, Hamburg 88.0 and Potsdam 87.4.
Paris took second place in Europe on 93.2. Barcelona scored 93.0, Florence 92.4 and Mallorca 91.6. Venice followed on 90.2 and the Algarve on 89.8. Lisbon and Vienna matched on 88.8 each, with Prague on 88.4.
Neumunster finished last in Germany on 27.8, with Ludwigshafen am Rhein on 28, Gera on 28.2, Solingen on 29 and Leverkusen on 29.4. Koszalin in Poland placed last in Europe on 51.4. Burgas scored 51.8, the south-west region of Ireland 53.2, Hauts-de-Seine 54.2 and Rimini 55.6.
Two events tables, one split result
The Bayreuth Festival tops the German event list on 95.2 points. The Landshut Wedding follows on 94.4, which MEININGER dates to 2027 because the city stages the pageant every four years. Leipzig takes 92.4 with the Wave-Gotik-Treffen, Kiel Week 91.8 and Cologne Carnival 89.4.
The Carnevale di Venezia leads the European event table on 95.4. Las Fallas in Valencia scored 95.0, the Fete des Lumieres in Lyon 91.4, Sziget in Budapest 91.0 and the Feria de Abril in Seville 90.4.
Oktoberfest carries two scores. The German events table lists 88.0 for the Munich festival and the European table lists 89.6. MEININGER scored the events in separate batches, so the two figures are two averages of the same five models. The release puts the festival at around six million visitors a year.
Where day-use inventory sits in the two leading cities
Dayuse lists day rooms in Heidelberg, the German city the models placed first. Its Heidelberg page carries seven day hotels across the city and its neighbouring towns, from $80.74 to $111.53 a booking, read 24 September 2026. Hilton Heidelberg sits at the bottom of that range and Leonardo Hotel Heidelberg City Center at the top.
Dayuse's Paris page, for the destination the models placed second in Europe, carries 50 day hotels from $69.70 to $226.54 a booking on the same read.
hotelSlots sells day rooms in four windows: morning 6AM-12PM, afternoon 12PM-6PM, evening 6PM-12AM and overnight 12AM-6AM. It states day-use rates run 30 to 75 percent below the nightly rate for the same hotel, counts 103,857 hotels across 46 U.S. states and charges no booking fee. Dayuse's Indianapolis page carries an average day-room price of $87.43, which is the platform's only published city average this run read.
MEININGER states the commercial case for the channel in its own release. The company writes that showing up in AI answers is becoming an economic factor for destinations and a way to reach new guests.
No publisher carries a day-room rate for Rome on a fetchable page, and the study carries no day-use data for any destination it ranks.
What the demand side measures
A Mower survey published on 21 September 2026 gives a measured read on how travellers act on an AI recommendation. Mower surveyed 1,000 United States leisure travellers aged 18 to 64 who had helped plan or book a leisure trip in the previous 12 months, on the Pollfish panel from 7 July 2026. Every respondent used AI tools at least occasionally, so the findings describe AI users rather than all United States travellers.
Twenty-eight percent of respondents said they would most likely begin planning their next leisure trip with an AI tool, against 30 percent who named Google search. Among AI users, 86 percent had used a tool to plan a trip and 97 percent of those said AI influenced at least one part of their most recent leisure trip.
Verification steps follow the recommendation. Fifty-nine percent of respondents searched Google, 49 percent visited an official website, 40 percent compared prices across sites and 20 percent asked a different AI tool the same question. Five percent booked exactly what the tool recommended. Thirty-four percent, the largest single group, trust AI only after verifying the recommendation. Twenty-nine percent said they mostly or completely trust AI to book a hotel and 28 percent said the same for a flight. About half would rely heavily on AI to plan a road trip, 14 percent would do so for a vacation costing $5,000 or more and 11 percent for a honeymoon. Thirty-three percent of parents of children under 18 picked AI as a preferred starting point for planning.
How to read the scores
Every AI score is the averaged output of five models on a date MEININGER does not state. The destination sets rest on 2024 overnight-stay volumes that the company attributes to Regionalstatistik and Eurostat, and no dataset accompanies the release.
MEININGER states the limits of the exercise itself. Malin Widmarc-Nilsson, VP Commerce at MEININGER Hotels, wrote that a top spot in the ranking does not mean every AI user will be pointed to those cities and sights. She wrote that recommendations differ noticeably between systems and that personal preferences feed into individual queries.
The German set is cities and the European set is regions. The release heads its European destination table City / region, and the methodology names regions. Both labels stand as published.