Insights
For Contributors
Share an article, perspective, or industry analysis for possible inclusion in the Alliance Insights section.
Featured Insight
Google has published an open specification — ARD — for how AI agents discover, verify and safely connect to tools across organisational boundaries. For hospitality, it is the standard that finally lets a booking agent find your property's live inventory without going through an OTA aggregator.
Read article
More Articles
By 2031, AI-driven automation will make most hotels operationally interchangeable. Terence Ronson argues the differentiator becomes the Human Experience Orchestrator — a role that reads AI outputs, applies behavioural psychology, and turns automated efficiency back into memorable hospitality.
Expedia and Booking are betting travellers will keep trusting them over hallucinating LLMs. Trip.com's James Liang is fighting a different battle — becoming the trusted inventory infrastructure the AI agents themselves rely on.
Consumer humanoid robots are entering homes and resetting guest expectations for relational AI. Terence Ronson maps the 2026–2031 hospitality response — on-device inference, transparent opt-outs, and a market bifurcating between human-centric sanctuaries and AI-native immersive stays.
Generic LLMs can't handle the operational complexity of a $10.9T industry — fare classes, rate codes, 20-30 disconnected hotel systems, seasonal reasoning. Abhijeet Patil argues domain-specific, fine-tuned models embed hospitality expertise inside model weights rather than bolting it on through retrieval.
Markus Busch argues every major chain's AI investment is designed for a guest who already showed up on the brand website — a customer profile that is disappearing. Discovery has moved to ChatGPT and Google AI, and only Accor has built anything meaningful outside its own walled garden.
Mario Gavira and Christian Watts brought their 18-month PhocusWire debate to the studio at Phocuswright Europe 2026. Gavira sees AI reinforcing the industry's largest brands. Watts sees it eroding their moats. The disagreement turns on one question: does loyalty still matter?
Hanna Schiller argues hotels are buying AI on top of a broken data foundation. Until guest identity is unified across PMS, CRM, loyalty and POS, AI will only scale the existing problems — inefficient marketing spend, weak loyalty economics, and personalisation that never quite lands.
Markus Busch draws a bright line between two AI timelines hotels are conflating. Internal AI (forecasting, pricing) moves on the hotel's clock. Distribution readiness for AI booking assistants moves on the market's clock — and that clock has already started.
At the Skift Data + AI Summit 2026 Mindtrip roundtable, travel leaders confronted the uncomfortable truth — AI assistants often hold richer traveler context than the hotels, airlines, and OTAs competing for the booking.
At TravelTech Show in London, Travelport CTO Andrew Jordan framed the shift travel is now making: from deterministic form-based search to probabilistic conversational intent. Panelists argued the biggest barrier to agentic commerce isn't trust or tech — it's consumer readiness.
Fred Bean argues hotel content is now competitive infrastructure. AI systems don't just index descriptions — they evaluate consistency across sources to decide whether a property is trustworthy enough to recommend. Governance, not publishing, is the new capability.
Markus Busch on the inversion of hotel distribution economics: AI agents run thousands of searches per session, and the cost of being "looked at" — long hidden inside OTA commissions — is shifting onto hotels going direct.
Tong Yin argues most hotel "AI pricing" is automated suggestion on three broken assumptions. Hotels override system recs over half the time and the system never learns — leaving 8–14% of revenue on the table in young markets.
Fredrik Sjoberg traces commission rates from 10% travel agents to 25% OTAs and warns the AI agent era could anchor a 30% "distribution tax" by 2030 — unless hotels move now to control their own AI booking interfaces.
Femke Nollet walks an independent hotelier through the AI booking journey, showing why API connectivity, hallucination audits, and MCP-readiness now decide whether a property keeps its 15–20% margin or hands it to an OTA.
Markus Busch argues the constraint on hotel success isn't AI capability — it's visibility. Over 90% of hotels are invisible to AI search, and most property-level ROI on tech investment is unmeasurable. Sight comes first.
Félix Pérez of Mirai breaks AI visibility into three layers — model memory, web search, and dynamic data via MCP — and explains which one matters at each stage of the guest journey, and where hotels should actually invest.
Dr. Tong Yin's contrarian read: properties with mature AI implementation are paying frontline staff 18–30% more, cutting attrition by a third, and reshaping roles around judgment and emotional repair — not replacing humans.
Hotels lead all travel sectors in AI machine-readability at 73% — and it doesn't matter. Markus Busch on why YouTube mentions (0.74 correlation) and editorial coverage drive AI recommendations, not structured data.
Shearly Reyes argues AI hasn't replaced search — it's extended it. With Google still at 70% share, search impressions up ~49%, and zero-click rates near 60%, hotels need SEO and answer-engine optimisation working together.
Are Morch flips the AI-adoption question: the bottleneck isn't access or budget, it's internal clarity. Unlike PMS or channel managers, AI's value is determined entirely by what hotels feed it — and most can't yet articulate it.
Glean's whitepaper makes the case that token cost is not a fixed expense but an architectural outcome. Retrieval, routing, memory, and model selection decisions compound across millions of interactions and determine whether enterprise AI is financially sustainable at scale.
OneTrust argues that AI governance committees create necessary alignment but cannot scale with enterprise AI adoption. Winning organizations are embedding controls into systems and workflows instead of treating oversight as a separate review function.
Terence Ronson warns hospitality not to repeat big tech's tokenmaxxing mistake — paying vendors by AI token consumption instead of guest outcomes. His proposed metric, Token Cost Per Guest, and six questions for the HITEC floor put operators back in control of the bill.
Lighthouse's Juanjo Rodriguez argues the bottleneck in hospitality AI is no longer insight — it's the gap between knowing and doing. Two-thirds of commercial leaders expect AI to reshape their business; only one in ten feel ready. The fix is AI that compresses the grind without taking the decision.
A Hotel Yearbook summary on why agent-to-agent commerce will change hotel distribution, shifting optimization from human browsing journeys to machine-readable trust, access, and fulfillment.
Senior commercial leaders are AI's slowest adopters — not for technical or organizational reasons, but because AI replicates the operational judgment their authority was built on. The fix is migrating their work upstream, fast.
Lighthouse analyzed 4,545 ChatGPT prompts across nine destinations. Most hotels never appear at all, a handful of chains and luxury properties dominate, and PR is now a distribution channel. The data on AI bias is measurable — and actionable.
Dr. Tong Yin argues hotels aren't failing on AI adoption — they're failing on architecture. Visible features without operational redesign produce "AI theatre." Revenue managers override pricing recommendations 50%+ of the time, and the override is the diagnostic.
Markus Busch dissects Booking.com's "protection from AI agents" messaging. The platform isn't shielding hotels — it's shielding its own seat between travelers and properties. Independents supply most of the room nights and can't see it from their own P&Ls.
Paid hotel CTR fell from 19.7% to 6.34% in roughly 15 months, mobile zero-click hit 77%, and Meta is overtaking Google in ad spend. Hoteliers need to rewire attribution, budgets, and content for AI-mediated search.
Perplexity and Harvard Business School analyzed real usage of agentic AI versus traditional search and found knowledge work is shifting from manual execution to supervisory oversight — with implications for how hospitality roles are scoped.
Version 2 of the AHLA/HTNG strategic brief introduces the DART taskforce's AI-Native Hospitality Commerce Ecosystem diagram — a five-layer reference architecture for how AI agents, hotel systems, and direct booking rails should interconnect.
Skyscanner CEO Bryan Batista wants AI agents to fix the broken handoff between metasearch and advertisers, and expects personal travel agents to reshape flight discovery. The implications for hotel distribution are direct.
Markus Busch argues Google's commission-free Universal Cart is a placement play, not generosity. By owning the AI surface where travel decisions get made, Google charges for *consideration* rather than transactions — shifting auction dynamics against hotels and OTAs alike.
Google's first year of AI Mode data shows travel queries have tripled in length — travelers are handing over paragraphs of intent, not keywords. Markus Busch argues a brief is not a long keyword; it's an instruction to an agent, and the hotel distribution stack just lost its unit of competition.
Pertlink's Terence Ronson proposes a new hotel KPI — Token Cost Per Guest — to measure generative AI spend per guest served. Drawing the parallel to how OTA commission moved from invisible deduction to board-level metric, he calls for USALI adoption before token costs sprawl unchecked across the P&L.
OpenAI's new Plaid integration is a blueprint, not fintech news. Hospitality needs the same kind of neutral, normalized infrastructure layer for AI distribution — or OTAs become the default operating system of AI-native hotel commerce by default.
Chris Anderson revisits his "billboard effect" research and argues AI is not just another visibility channel. It interprets traveler intent, curates a small choice set, and dynamically represents properties — shifting the competitive unit from inventory presence to semantic fit.
Google rewired Search, Universal Cart, and launched Gemini Spark at I/O 2026 — collapsing the dream-plan-book funnel into a single agent-mediated layer. Hotels not exposed through MCP risk being invisible to the agent doing the booking.
Google has published its first official guidance on optimizing content for AI search features — and the headline is that traditional SEO fundamentals still do most of the heavy lifting.
Melissa Jurkoic argues hospitality's biggest barrier to AI adoption isn't technology — it's trust. Without trust readiness, even the most mature AI roadmap stalls, and AI risks eroding the very relationships the industry depends on.
Mario Gavira maps every major travel category onto an Agentic Disruption Risk Matrix and finds the picture is more nuanced than the hype suggests: human-led players are safest, the largest cluster sits behind moats of loyalty, unique inventory and bundling, and only a small cohort — second-tier OTAs, low-cost carriers and commodity independent hotels — faces real structural exposure.
An interactive map of how AI-driven demand, orchestration protocols, data flows, and hotel supply systems connect in the new era of hospitality distribution.
In a Hospitality Net opinion piece, Ira Vouk argues that the most important AI shift in hospitality is not another chatbot but the rise of embedded digital employees that can execute cross-system workflows inside hotel operations.
In an AltexSoft interview, Ira Vouk argues that MCP can improve hotel AI discoverability and integration, but it will not solve hospitality fragmentation on its own. Adoption, data quality, and merchant-of-record control remain the harder problems.
George Roukas's argument that the real disruption in travel is not chatbot UX but the quietly advancing infrastructure behind agentic discovery, orchestration, and booking.
A plain-language guide for hoteliers on the difference between AI assistants and AI agents, how to build a first digital employee, and where agentic workflows will matter most in hospitality.
Most hotels believe they are engaging with AI. Michael Goldrich argues they are experimenting — and that there is a critical difference. True AI advantage requires crossing three simultaneous thresholds: literacy, workflow, and discovery.
A clarification of the three fundamentally different interpretations of "agentic hotel booking" - AI-assisted, AI-mediated, and AI-executed bookings.
A report on which hotel brands and properties are most visible in AI-generated search results across ChatGPT, Google AI, and Perplexity.
A practical white paper on how hotels need clear, consistent, machine-readable data to stay visible as AI assistants reshape travel discovery.
78% of hotel chains deploy AI, yet only 7% have a comprehensive AI strategy. Michael Goldrich maps what actually worked in 2025 — voice AI, MCP infrastructure, agentic booking engines — and what the 2026 shift from generative to agentic AI demands.
Why hotels should stop treating AI crawlers as a nuisance and start seeing crawlability as a direct-booking visibility issue in the age of zero-click search.
AHLA/HTNG STRATEGIC BRIEF: Leveraging MCP, Hotel Content, and AI Tools to Maximize Direct Bookings VERSION 1 | PUBLISHED: JANUARY, 20, 2026 This document was developed by the Global Technology 100.