Destination AI Forum 2026: What Was Said on Stage

In my personal article I wrote about what stayed with me from Destination AI Forum. In this piece I want to share what was actually said on stage. Destination AI Forum 2026 took place on 29 and 30 September at the National Housing Center in Washington, D.C. It was the third edition, it was sold out, and there were more than 500 people from over 280 companies.

Something that came back in a lot of sessions: nobody really knows yet where AI is taking our industry. Harman Singh Narula of Canary Technologies said he has no crystal ball, and Andrew Boch of Mobi.AI said there is no rule book yet. What has changed is how many hotels are now using AI. Anil Aggarwal of Milestone put it well on day two: nobody asks anymore whether to do this. They ask how.

Where hotels stand today

The event started with the The State of AI in the Hotel Industry study, presented by Josiah McKenzie of Hospitality Daily and based on answers from more than 100 hospitality leaders. Nine out of ten said AI already makes their routine work faster, and nobody plans to spend less on AI in 2027. Most of them feel excited and curious about it, although the most common negative feeling was being overwhelmed.

Later that day, in his own session, Narula shared an interesting number from Canary's data. Of tens of millions of guest messages, only about a third are questions, such as what time breakfast starts. The rest are requests, such as a late checkout or a shower that does not work. Most AI tools today can answer the questions, but the next step is AI that also handles the requests itself, for example by arranging the late checkout or reporting the repair. Narula expects that to grow quickly over the next 6 to 18 months.

What is working today

Most examples with real numbers came from the back office.

• Food and beverage. Nancy Wolff of Columbia Hospitality shared how an AI tool read an operator's invoices and found that a price change on salmon was costing about 2,900 dollars a month.

• Procurement. HHM Hotels and Folio found about 370,000 dollars a year in potential savings, with software that suggests a cheaper approved product while ordering.

• Hiring. Commonwealth Hotels and Ponte screened more than 2,400 applicants by phone in three months, and saved 685 manager hours across eight hotels.

In her session, Wolff made a point I liked. The AI projects that work best often do not start with the IT team or a big new system. They start with the people in the hotel who deal with the problem every day. Her advice to technology leaders: listen to those people, and give them the freedom to try things themselves.

Dan Kornick, CIO of Loews Hotels, was also very clear about how he decides. If an AI feature costs more than the people doing the job today, he does not buy it.

What is coming in 2027

Distribution came up in many sessions, and one term I heard again and again was AI visibility. In other words, does an AI assistant name your hotel when a traveller asks where to stay?

Chris Anderson of Cornell and Nick Slavin of Curacity shared research showing that about 85 percent of hotels never show up in an AI answer. So how does AI decide which hotels to recommend? Not so much by what the hotel says about itself on its own website, but by how often other trusted sources talk about it, such as travel articles, guides and reviews. In their study, the hotels that were mentioned most often by those sources were recommended about three times as often as the hotels that were mentioned least.

The good news is that hotels can do something about it. Slavin shared the example of The Kahala, a luxury resort in Hawaii. It worked on getting mentioned more often on trusted outside sites, and is now recommended five times as often as the hotel with the fewest mentions in its competitive set.

Being recommended is one thing, but AI also has to get the facts about your hotel right. That is where your own website does matter. Anil Aggarwal of Milestone shared how AI tools kept saying that a boutique hotel in New York offered room service, years after it had stopped. One new page on the hotel's website fixed that within 30 days.

Greg Land of AWS told hotels to rewrite their 2027 distribution strategy now. If they do not, he said, AI platforms will just connect to the existing intermediaries. At the same time, real bookings through AI assistants are still rare. Wyndham and Mobi.AI, who built one of the first hotel booking tools inside an AI assistant, said openly that adoption is still low, and that the value so far is mostly in what they learned.

What is holding hotels back

The technology itself is rarely the problem. In the study, only 8 percent blamed its limits. Three other things came up again and again.

• Data. Only 42 percent of leaders feel confident their data is ready. Mike Baldinger of Evention gave a simple example: the same transaction code can mean something different at two hotels in the same group.

• The frontline. 84 percent of leaders said their own work is getting better, but only 15 percent see a difference for their frontline teams. Shane O'Flaherty of Microsoft called frontline workers the forgotten people of this technology.

• Checking the work. Kelly McGuire of Kasa found an owner report in which AI had given the credit for a revenue gain to the wrong sales channel. Her rule: if AI saves you ten hours, spend five of them checking the result.

Training was the answer that came up most. Keryn McNamara, CIO of Aimbridge Hospitality, said what keeps her awake at night is no longer cybersecurity, but whether her associates are fluent in AI.

The start-up competition

Day one also had the first hospitality AI start-up competition, with six young companies pitching their ideas. They ranged from a tool that helps guests with dietary needs find what they can safely eat, to a voice assistant that lets guests book directly on the hotel website. StayFull, an AI-native property management system for small independent hotels, won both the judges' and the people's choice.

Should versus could

Jim Harding, CEO of Otonoma, closed day one with a different question. After a day full of adoption stories, he asked whether we should, and what we need to protect. He told about an Amazon recommendation engine that was rushed to market and had to be taken down after it hurt some customers. His advice: test more than feels necessary.

The AI Showcase

On day two, a new format was the AI Showcase, where a supplier and a hotel company shared the stage for seven minutes, followed by questions from the room. Next to HHM with Folio, Commonwealth with Ponte and Wyndham with Mobi.AI, there were two more pairs. Accor and Azira ran a targeted campaign for the relaunched Sofitel New York, aimed at travellers who are not loyal to one brand. DiamondRock Hospitality and PepperMill connected the systems at Cavallo Point, so department heads get a daily suggestion for which shift to swap to avoid overtime.

The Great AI Debate

One of the most fun sessions put three PMS leaders on one stage: Adam Harris of Cloudbeds, Jacob Messina of Stayntouch and Richard Valtr of Mews, with Sloan Dean of AI Hospitality Group as moderator. The big question was whether hotels should run on one platform that does everything, or pick the best system for each job. Harris argued for one unified platform, because value gets lost in the gaps between systems. Messina prefers a focused PMS with strong partners, so hotels can replace the parts that do not work. Valtr was somewhere in the middle, with the core apps built in and an open system for the rest. They also disagreed on how hotel data should be used to train AI. At the end, the audience voted Jacob Messina the winner.

Can AI deliver hospitality?

Right after the debate, a panel led by Esther Hertzfeld of Questex looked at the guest experience. The message was that AI often works best when the guest does not even notice it, but only if the hotel really knows who the guest is. Branigan Mulcahy of Virdee told about a luxury hotel that hosted a one million dollar wedding. A year later, the couple came back for their anniversary, but this time they booked through an OTA. Because their guest profiles were not linked, the front desk asked them if it was their first stay. For me, that story says more about the importance of good guest data than any number.

The big brands

Day two ended with the big brand panel, led by Stuart Greif of Forbes Travel Guide, with leaders from Hilton, Hyatt, Marriott, Wyndham, Choice and IHG. It was the third year in a row that this panel came together, so you could really see how much has changed. A year ago it was mostly about testing and training. Now AI is really in use at all of these companies. Pat Nestor of Hyatt said the focus has moved from adoption to absorption: are people really getting value out of the tools they already have? Lindsay Liszewski of Choice said the challenge is no longer getting people to use AI, but choosing what to do first. And Nitin Sood of Marriott described their approach as tech-enabled, human-centered. For me, that was a good way to end the two days.

For what stayed with me personally from these two days, read my article: https://www.hospitalitynet.org/opinion/4134972/destination-ai-forum-2026-hotels-are-ready-now-comes-the-how.