I let Claude, ChatGPT, and Gemini plan my entire vacation, and the results weren't what I expected

Published Aug 9, 2026, 6:00 AM EDT Mahnoor Faisal is a tech journalist covering AI and productivity tools with bylines at XDA, SlashGear, MakeUseOf, Laptop Mag, and Android Police. She's been writing professionally since she was sixteen, and has since penned hundreds of articles. This includes in-depth coverage of AI tools like NotebookLM to breaking news across the AI space. Her passion for technology started when she received her first iPod Touch (4th generation) on her 8th birthday, and she's been deep in the tech world ever since. Currently pursuing a degree in computer science, Mahnoor brings both a journalist's eye and a technical foundation to her coverage of how AI is reshaping the way we work and learn. I've been writing about AI tools since before ChatGPT became a household name, and something I've noticed since then is that one use case the labs building these tools keep circling back to is travel. It's a trend I've seen across demos, keynotes, press releases, and even the casual examples they drop on social media. I've tried using the same scenario to put models to the test, but my testing has always been superficial, since I was always testing a made-up trip I had no intention of taking. However, I finally took a much-needed vacation this summer, and I figured there was no better time to stop testing hypotheticals and let Claude, ChatGPT, and Gemini plan a vacation I was actually going on. So that's exactly what I did. Claude was the best at keeping my preferences in mind The advantage of an assistant that already knows you To preface this, Claude's been my go-to AI chatbot for a while now, and it's the one I rely on most. This means I spend a lot of time with it, and it knows my recent projects, my habits, my quirks, and whatever else I've fed it over months of conversations. Given this was a trip I was actually going on, this extra context did matter, so I didn't bother stripping that context out to keep things even. It gave Claude a bit of an unfair advantage over ChatGPT and Gemini, but it also reflects how I'd actually use it. That said, throughout my trip planning, I found that Claude was the best at keeping my preferences in mind. Beyond my broader preferences, dietary restrictions, and budget, it also remembered the smaller stuff I threw around throughout the planning — like the fact that I prefer a slower pace, that I'd rather have one relaxed sit-down meal than three rushed ones, and that I wanted a coffee or matcha stop worked in somewhere. ChatGPT was fairly good about this too, though I noticed it would take a bit longer to update its memory summary, and I would sometimes need to repeat myself before it actually registered a change. Gemini wasn't very good at this, and it kept forgetting even the basics. For instance, I decided to take this specific trip completely solo, which meant the attractions I was looking at were ones that made sense for someone traveling alone. But Gemini repeatedly defaulted to group and couple-oriented recommendations, and I'd have to remind it that it was just me. Why would I want to do a romantic candlelight dinner, Gemini? What I really liked about using Claude for my trip planning was that it was more willing to push back than the other assistants. Gemini has a habit of agreeing with you, and there are times when GPT's models also fall into the same trap. I noticed both assistants telling me my plan was great when it could clearly use some work. Claude, on the other hand, wasn't afraid to tell me when I was wrong. When I tried to cram too much into a single day, it flagged that the plan was unrealistic and walked me through why. Claude also felt the most conversational out of the three. It'd suggest places that actually felt like me, and instead of just listing them, it would ask a follow-up or two and shape the rest of the plan around what I said. Overall, I found Claude to be a good starting point for the whole trip. It seemed to understand the messy bits of my trip and helped me smooth out the parts I hadn't figured out yet. ChatGPT was the reliable all-rounder Master of none, but I didn't mind It's been a while since ChatGPT's impressed me, but my trip planning showed me why it's still the one so many people reach for first. It wasn't necessarily the best at any one part of planning my vacation, but it was consistently good at almost everything I threw at it. That ended up being more useful than I expected. Vacation planning isn't really one task. One minute, I was comparing neighborhoods to stay in, the next I was trying to figure out whether two attractions made sense on the same day, and five minutes later I was asking how much money I should realistically set aside for food and Grab rides. ChatGPT handled those jumps without much trouble. Where I particularly liked ChatGPT was once I already had the bones of my trip figured out. Instead of constantly asking it to regenerate an entire itinerary, I could make tiny changes: move one attraction to another day, swap a restaurant out, add a shopping stop, or ask whether a particular day looked too packed. It was also good at giving me options instead of acting like there was one objectively correct way to spend my vacation. If I wasn't sure what I wanted to do one evening, it'd usually give me a few directions I could take depending on whether I wanted food, shopping, sightseeing, or just an early night. That made it useful for the parts of the trip where I hadn't made my mind up yet. Still, it had some of the same problems I ran into with Claude, only more frequently. ChatGPT could be a little too eager to make whatever I suggested work. If I asked whether I could squeeze another stop into an already packed day, there were times when it would happily rearrange everything instead of simply telling me that I was doing too much. That's not exactly what you want from something that's supposed to help you avoid turning your vacation into a military operation. The same experience as above showed when I leaned on ChatGPT to plan an upcoming work trip based on what I had already booked. ChatGPT managed to turn the trip, which has admittedly become a bit of a mess, into something coherent. It picked up the events already sitting on my calendar, noticed where some of them overlapped, and pointed out which commitments I'd realistically have to skip or move. Instead of simply dropping everything into an itinerary and pretending it would all work, it separated the busier work days from the parts of the trip where I'd actually have time to explore. Claude failed at the same task. Despite being the assistant that knew me better overall, it pulled in an older departure date and consequently underestimated how much accommodation I still needed. Claude was excellent at retaining the softer context surrounding my trip. It knew how I like to travel, what I enjoy, and the sort of itinerary I'd actually want. On the other hand, ChatGPT was better at reconciling the decidedly less glamorous stuff: dates, bookings, calendar events, conflicts, and gaps that still needed filling. The latter is what matters more when you're actually trying to turn a pile of reservations and commitments into a trip that works in the real world. Gemini had an advantage the others couldn't really match It peaked the moment I landed Gemini was easily my least favorite of the three when it came to actually planning the trip. As I mentioned earlier, it wasn't particularly good at keeping track of my preferences, and I found myself repeating things I'd already told it far more often than I did with Claude or ChatGPT. But once I was actually on vacation, Gemini suddenly became a lot more useful. At one point, I was standing at a major transit station trying to figure out how to get to another part of the city. I could have opened Google Maps myself, of course, but instead I asked Gemini. It immediately pulled up the public-transit route through Google Maps and walked me through which train line to take, which direction to travel in, how many stops I'd be on it for, and what I'd find when I got off. That sounds like a small thing, but it's exactly the kind of question you end up asking constantly while traveling. Unlike a generic AI-generated answer about how public transport usually works in a city, Gemini could surface the actual Maps route alongside its response. I noticed the same advantage when I asked it to suggest hotels near a particular train station. Gemini didn't just give me a list of names based on what it knew. It surfaced Google Hotels results directly in the conversation, complete with ratings, nightly prices, and links to see each property. This was where Google's ecosystem gave Gemini an advantage that neither Claude nor ChatGPT could really replicate in quite the same way. Google already has Maps, Hotels, Flights, and Search, and Gemini can pull those services into what would otherwise be a normal chatbot conversation. Instead of using the AI to make a plan and then opening another app to check whether that plan made sense, I could sometimes do both at once. The best travel planner was all three Ultimately, every assistant I used to plan my vacation had its strengths and weaknesses. Claude was the best at understanding the kind of trip I actually wanted and keeping my preferences in mind, while ChatGPT did a better job of keeping track of the logistical mess that comes with planning a real trip. Gemini was the weakest of the three when it came to remembering context, but its integrations with services like Google Maps and Hotels became genuinely useful once I was actually traveling. If I had to plan another vacation tomorrow, I don't think I'd rely on just one of them. I'd probably start with Claude to figure out the shape of the trip, use ChatGPT to make sure all the moving pieces actually fit together, and keep Gemini around for the inevitable moment when I'm standing in an unfamiliar train station wondering where I'm supposed to go.