AI in Travel: All Talk & No Data

The more you look at how OTAs and the wider industry are using AI, the more obvious it becomes; they’re not trying to solve real problems. They’re using AI to dress up tired old data, calling it innovation, and hoping no one notices.

Instead of solving real problems, they’re rewording guest reviews, reprocessing old information, tweaking booking flows, and pretending that predictive recommendations are groundbreaking while they narrow our choices and strip away the joy from our trip planning.

Nobody wants a chatbot or agent to decide where they will stay during a trip, and stealing the joy out of the discovery process isn’t a smart long-term move for the industry.

Travellers enjoy discovery; they love looking at and choosing hotels while planning their trips. It is, by nature, an emotional process more than a logical one, which is why people prefer to choose hotels on Instagram rather than ChatGPT.

Underserved Traveller Segments Are Ignored

The exception to all of this, where AI could be useful, is for travellers with hard requirements, like disabled travellers or dog owners, who need to know if a hotel will meet their needs before they book. These two groups are what the industry calls underserved market segments, and they are underserved because we don’t have the data to match them with hotels that genuinely meet their requirements.

It’s a traveller pain point that OTAs like Booking and Expedia and hotel certification bodies completely ignore. Instead of verifying which hotels are genuinely accessible or pet-friendly, they slap labels on properties based on unchecked self-reported data.

This leaves disabled travellers and pet owners to find out the hard way, when they show up and discover that "accessible" means "one ramp somewhere in the building" or that "pet-friendly" means "dogs allowed, but only for a fee and not in any of our public spaces."

Verifying Claims Is Priceless

I had a great conversation with Alvaro Silberstein , the founder of Wheel the World, a specialist OTA for disabled travellers that exists purely because of how unreliable accessibility information is in mainstream travel booking. He told me that since there’s no universal standard for what "disabled-friendly" means, the term gets abused by hotels trying to market themselves as inclusive, whether they actually are or not.

So Wheel The World verifies disabled friendly hotels— checking everything from doorways to shower access and verifying their claims because nobody else has bothered to check.

The same problem exists for pet-friendly hotels.

The OTAs blindly accept whatever a hotel claims, meaning dog owners are left sifting through vague reviews, hoping they don’t end up in a place that bans dogs from public spaces, or only allows toy breeds. AI could fix this by proactively gathering and verifying hotel data, but instead, the industry uses it to regurgitate guest experiences.

OTAs excel at directing generic travellers to hotels ranked by guest reviews but completely fail to match underserved travellers with hotels that meet their needs. It’s partly because they don’t have the data but mostly because they never bothered to ask the right questions.

AI Is Great At Asking Questions

Instead of using AI to churn out chatbot scripts and tweak booking algorithms, the right way to use it is to gather granular data that doesn’t already exist—the kind of useful data that solves real-world problems for underserved traveller segments.

Right now, no AI chatbot can tell you how friendly a hotel really is to disabled guests or dog owners because that data doesn’t exist in a structured, verified way.

AI can’t magically answer questions about a hotel’s pet policies, accessibility, or amenities if nobody has collected that information in the first place. But if that data did exist, if an AI chatbot actually had verified, structured hotel data, then it could be genuinely useful, helping travellers in these underserved segments find hotels that fit their needs.

If this was the case, people might actually use a chatbot to book a trip...

The Real AI Opportunity Is Data

The real opportunity isn’t in chatbots or predictive booking flows. It’s in data.

For dog owners, disabled travellers, and anyone with specific travel needs, that data makes all the difference. Academic research tells us that dog owners value granular and accurate information about pet-friendly hotels before they book a room, but the industry has failed to deliver, to the point that dog owners no longer trust the words pet-friendly.

Roch Dog steps into this gap, using AI not as a gimmick but as a tool to gather granular, verified data that actually matters. With every hotel we certify, and every piece of dog-friendly information we verify, the closer we are to serving a real market need.

We use what we KALI (K9 Aligned Lodging Inspector) Agents to proactively contact hotels and ask them lots of questions in order to gather lots of data about their hotels.

We don’t just collect this data; we process it through an NLP pipeline, first using contextual analysis, then comparative analysis, and finally cluster benchmarking to rank and review hotels algorithmically. This lets dog owners make informed decisions and know which hotels genuinely welcome dogs and which slap "pet-friendly" on their website.

No OTA is doing this. They don’t know which hotel is actually the best for pet owners because they never bothered to collect the right data, and even worse, they are giving pet-friendly hotel awards to terrible pet-friendly hotels, setting dog owners up for failure.

The demand already exists. The spending power is there. The only thing missing is a platform that actually understands dog owners and their needs.

That’s why Roch Dog isn’t just another AI startup; it’s the only solution that gives dog owners the confidence to book with certainty and plan their trips with confidence.

Once you own the data, you own the market.