Hilton wants to be valued like a tech company because it says it is plugged into every AI booking agent. Fine. But a Nasdaq listing and an Uber CTO prove nothing. The only test that matters is whether the money saved on distribution ends up with the hotel owners, or stays with Hilton.
That is the question hoteliers should be asking after Hilton’s third-quarter earnings call, as reported by Skift. And right now, nobody outside Hilton can answer it.
What Hilton actually said
Chief executive Christopher Nassetta now describes Hilton as “people-led, tech-powered”. In the past month the company has hired Jeff Moore, who led a 2,500-person engineering team at Uber, as chief technology officer. He reports straight to Nassetta. Hilton has also said it will move its stock listing to the Nasdaq, home of Apple, Microsoft and Nvidia, while insisting the move changes nothing operationally.
On the call, Nassetta said Hilton is working with “each and every one” of the AI agents emerging in travel. He pointed to a direct integration with corporate travel platform Navan, launched in July, that skips the global distribution systems. He also leaned on Hilton’s scale: roughly 25% of quality US hotel supply, according to Skift’s summary. And he said lower distribution costs would eventually flow back to owners.
Notice the word “eventually”.
Four weeks ago, the story was very different
Earlier this month I wrote that Hilton’s AI agent problem was a gift to the OTAs. In Skift’s own tests, Meta’s Muse agent hit a “Something went wrong” error on a Hilton booking page. A rival agent worked fine. The point was simple: if an agent cannot get a price, the guest books somewhere else.
So “we work with every agent” is a big claim to make a few weeks later. Skift itself asks how much of it is live integration and how much is intent. That is the right question. Being “connected” to an agent can mean anything from a full booking API to a partnership email. Investors are being asked to price in the first. Owners should insist on proof of the second.
The real prize is cheaper distribution
Here is why this matters more than the Nasdaq headlines. Every booking that comes through an OTA or the GDS carries a cost. If an AI agent can book Hilton directly, a chunk of that cost disappears. That is real money, and it is the most interesting thing AI could do for hotels.
But the savings do not land in one place by default. They could go to Hilton through its fees and its share price. They could go to owners through lower charges. Or, more likely than anyone admits, they could go to the agents themselves. Meta is already weighing a percentage cut of travel bookings made through Muse. A direct connection to an agent that takes a commission is not the end of distribution costs. It is a new tollbooth with a nicer interface.
Hilton is courting owners through RISE, its fee-discount scheme tied to guest satisfaction scores, with a RISE 2 focused on tech and AI efficiencies due in the first quarter. That will be the moment to judge this. If RISE 2 hands owners a clear, measurable cut in distribution cost, Hilton’s pitch is real. If it is vague, the tech story is mostly for shareholders.
Owners are already paying more to sign up
There is another number in the same call that should make owners pause. Skift reports that key money, the cash brands pay owners to sign deals, is now involved in about 11% of Hilton rooms under construction, up from 9% to 10% five or six years ago. Hilton also had its biggest quarter of organic openings on record, with 27,000 rooms.
Brands are competing hard for owners. That gives owners leverage. They should use it to ask a blunt question: when an AI agent books my hotel directly, what do I pay, and how is that lower than an OTA booking? If the brand cannot answer with a number, it is not a tech platform. It is a hotel company with a better slide deck.
Scale is Hilton’s weapon. Independents need a different one
Hilton can get direct deals with agents because it controls a quarter of quality US supply. An agent that ignores Hilton has a gap. An agent that ignores a 40-room independent has no gap at all. That is the uncomfortable truth behind this story.
Big brands will get agents to come to them. Independents will have to be found. That means clean, accurate, machine-readable content everywhere AI looks, which I covered when I argued that AI isn’t snubbing independent hotels, it just hasn’t heard of them. It also means treating OTA listings as sales copy, because Booking.com’s AI is already answering guests from your listing.
Wyndham has a more honest take on all this. It is treating AI agents as a new type of guest rather than calling itself a tech company. Hilton has gone further and louder. Louder is fine, as long as the results follow.
My take
Hilton is right that direct connections to AI agents are where hotel distribution is going. It may well be ahead of its rivals. But a share listing does not book a room, and a CTO hire does not cut a commission. Judge Hilton, and every brand making the same pitch, on one thing: published, falling distribution costs for owners. Until then, “tech-powered” is a valuation story, not a guest story.
So, owners and revenue leads: would you rather have a brand that talks to every AI agent, or one that can show you what each agent booking actually costs you?
Want more takes like this on AI, travel and how brands get chosen? Subscribe to my LinkedIn newsletter and join the debate.