The Big Picture
TLDR
‣ Travelers adopted AI roughly two years ahead of the travel industry, and that gap now shapes who gets booked.
‣ Generative AI lifted novice productivity 34% while experienced staff barely improved, narrowing the gap between junior and senior revenue managers.
‣ 15,976 hotel projects are in development, with 7,800 openings by 2028, entering a market where AI assistants route demand.
Somewhere right now a hotel group is signing off on a 1,000-room resort while the layer that decides who fills those rooms is being rewritten by systems nobody in that room controls.
That's the tension this week. Capital is committing to physical supply on a five-year horizon. Distribution is mutating on a five-month one.
The Mastercard Economics Institute frames 2026 travel around three forces: macro conditions, machines, and motivation. Record passenger volumes, longer trips, new destinations rising in the rankings. Easing inflation and falling rates are giving people permission to spend again. The machines part is the one most operators are still treating as a side project.
Brief of the Week
Travelers are running roughly two years ahead of the travel industry on AI adoption. That gap is the whole story this week.
Read the full brief on TourismIntel →The Number
15,976
According to Hosteltur, there are 15,976 hotel projects currently in development worldwide, with over 7,800 openings expected by 2028 and the US and China accounting for nearly 60% of them. This means a wave of new rooms will hit a market where the demand routing layer, AI assistants, social feeds, agentic booking, looks nothing like the one those pro formas assumed.
01
Social feeds are now a booking channel
According to Hosteltur, over 43% of travelers have been influenced by a viral post or video when choosing a trip, concentrated among Millennials and Gen Z. This means the inspiration phase has already left your website and your OTA listing. The cost is control: you're being sold by people you don't brief and can't correct.
02
AI compresses the skill gap, not the talent gap
According to Silicon Canals, one study on generative AI in the workplace found productivity jumped 34% for novice workers while experienced employees improved barely at all. This means the junior revenue manager with good prompting habits now performs close to the senior one. For hotel groups scaling into 7,800 new properties, that's the staffing model changing under your feet.
03
Álava builds the machine-readable layer
The Diputación Foral de Álava has activated Aktibatu Araba, connecting local operators to Spain's Intelligent Destination Platform under the national recovery plan. Boring on the surface. It's the only story here where a destination is structuring its data so machines can read it. Most DMOs are still producing PDFs.
04
La Rioja shows the other side
According to Hosteltur, illegal accommodation supply and touristification are damaging tourism's reputation in La Rioja. Two destinations, same country, opposite trajectories: one is instrumenting itself, the other is losing consent from residents. Neither problem gets solved with a campaign.
Contrarian Take
Flip the camera. Stop looking at the pipeline from the investment committee and look at it from a traveler asking an AI assistant for three nights near Rioja wine country in October.
That assistant does not see your 1,000-room resort's brand equity. It sees structured data, availability, verified reviews, price, and whether your content is machine-readable. If it isn't, you're invisible, no matter how much concrete you poured.
My position: the biggest risk in hospitality right now is not oversupply. It's supply built by companies that invested in real estate and skipped the data layer entirely. Álava, with a fraction of the budget, is better positioned than most chains.
Where This Goes
Pick one property. Take your top ten booking queries and run them through ChatGPT, Gemini, and Perplexity this week. Write down what comes back and who owns the answer. If it isn't you, you know exactly what to fix before the next capex cycle.