The Hotel Industry Has Been Waiting for AI to Save It From the OTAs.
AI Is Not the Second Coming. It's the Second OTA.
Argues that AI will strengthen rather than disrupt OTAs like Expedia, as these platforms already control the massive datasets and connectivity that power AI recommendations.
Photo by Americas Great Resorts
The Hotel Industry Has Been Waiting for AI to Save It From the OTAs.
Every conference. Every keynote. Every vendor pitch deck.
AI is the savior. AI will disrupt the intermediaries. AI will give hotels back their guests, their margins, their dignity. The faithful are on their feet. The collection plate is full.
There is only one problem with this sermon.
The god they are praying to works for Expedia.
The Industry Has Held This Revival Before
It was the early 2000s. A new technology arrived that promised to put hotels in front of more travelers than they could ever reach alone. All you had to do was list your inventory. The platform would handle the rest. Discovery. Comparison. Booking. Done.
The industry celebrated. Finally. A solution.
Two decades later that solution extracts 15 to 25 percent commission on rooms sold through it, a range consistent across reporting from Phocuswright, Skift, and hotel operator disclosures. It mediates the relationship with the guest. It controls the ranking. It sets the visibility rules. And it changes those rules whenever it wants to.
The hotels that listed their inventory in the early 2000s did not gain distribution. They outsourced their demand. The real economics of OTA dependence have compounded against independent properties for two decades and show no sign of reversing.
The hotels currently celebrating AI liberation made the same bet once before. The sermon is identical. Only the platform changed.
Meet the New God. Same as the Old God.
Here is what the AI salvation narrative requires you to believe.
That a technology running on data will somehow disadvantage the companies sitting on the largest travel data sets in human history.
That a technology requiring inventory connectivity will somehow bypass the platforms with electronic connections to hundreds of millions of accommodations worldwide.
That a technology built on behavioral signals will somehow ignore the companies with decades of transaction history, conversion data, and traveler preference profiles that no independent hotel could assemble in ten lifetimes.
Booking.com does not have a data problem. Expedia does not have a connectivity problem. They do not have an engineering problem or an AI readiness problem or a structural vulnerability that a language model is going to exploit on behalf of a 60-room boutique resort in the Hudson Valley.
They have a head start measured in decades and a budget measured in billions. The luxury hotel marketing agency that tells you otherwise is selling you the sermon, not the solution.
AI does not erase those advantages.
AI is those advantages, industrialized.
How AI Actually Reads Your Hotel
AI travel recommendations do not begin with hotel quality. They begin with available signals.
When an AI system is asked to recommend a luxury hotel in a competitive market, it tends not to surface the best property. It surfaces the most legible one. The property with the broadest, most consistent, most machine-readable footprint across the open web. Reviewed extensively. Described consistently. Present across every major platform the model was trained on.
Expedia Group has publicly stated it holds more than 70 petabytes of historical and real-time travel data, with hundreds of AI models making more than 900 billion predictions across its ecosystem annually. That is not evidence of OTA vulnerability to AI disruption. That is a description of structural AI advantage already in place.
A 2026 Skift analysis found that AI travel responses often cite third-party comparison and review sources ahead of direct brand websites. When an AI system was asked about a specific Hyatt property, the source it cited most was not Hyatt. It was NerdWallet. That is the point hotels keep missing. AI does not begin with brand preference. It begins with retrievable evidence. The property with the broader, more consistent, more machine-readable footprint gives the system more confidence than the property with the better story but the weaker signal base.
AI is not a bypass. It is an amplifier. And it is amplifying exactly what already existed.
The Wolf Built the Church
While the industry has been holding revival meetings about AI liberation, the platforms have been building.
Booking.com has embedded AI across its ranking and recommendation systems. Expedia has deployed AI-driven personalization at a scale that independent hotels cannot approach. The OTAs are not victims of the AI revolution. They arrived at it first, with more data, more connectivity, and more engineering capacity than the entire independent hotel sector combined.
The independent hotel with 40 rooms and a spa does not have a training data set. It does not have behavioral profiles on hundreds of millions of travelers. It does not have the infrastructure to compete with platforms whose entire business depends on predicting which property closes the booking.
The wolf did not come to the door.
The industry built the church and sent out the invitations.
The platforms did not arrive late to AI. They arrived early, with the data, connectivity, and transaction infrastructure that hotels still do not control and cannot replicate.
The Second OTA
AI Is Not the Second Coming.
It's the Second OTA.
The hotels that listed their inventory two decades ago because they believed the platform was their partner did not get partners. They got landlords. The hotels celebrating AI as their liberation technology in 2026 are making the same category error with better graphics in the pitch deck.
The platforms that extracted two decades of margin from independent hospitality did not get disrupted by AI. They got upgraded by it.
The independent hotel that goes into the next decade with its demand still rented from intermediaries, still dependent on OTA visibility, still without a direct relationship with the travelers it needs, will discover that AI made its situation structurally worse, not better. What luxury hotel marketing agencies actually do in this environment looks nothing like what most properties are currently being sold.
The collection plate is still full.
The god they are praying to just got a lot more powerful.
Comments
Comments for this content
0 comments available