Tourism is entering an inflection point that feels familiar to anyone who has watched other industries digitise at scale. Artificial Intelligence is no longer experimental. Planning is smarter, personalisation is deeper, and operations are increasingly automated. And yet, for travelers, the overall experience often feels more fragmented than ever.

The problem is no longer intelligence. The problem is coordination.

Airlines, airports, hotels, destinations, mobility providers, and regulators are each deploying AI within their own domains. But tourism is not a collection of isolated transactions; it is a living system that only works when decisions taken in one part of the journey are understood and acted upon by all others. Today, that coordination breaks down precisely when autonomy increases. This is the paradox facing global tourism.

A delayed flight still triggers a cascade of manual interventions. The hotel is unaware. Transfers are not adjusted. Reservations are missed. The traveler becomes the system integrator, making calls across time zones, languages, and platforms. This happens millions of times a day, across every market. Not because technology is lacking—but because there is no shared coordination layer.

That gap is what Agentic Tourism is designed to address.

Launched in partnership with TOURISE and the Saudi Ministry of Tourism, the Agentic Tourism Initiative introduces a new operating model for the visitor economy: one where autonomous AI agents across airlines, airports, hotels, destinations, and experiences can coordinate in real time, across borders, under clear human governance. It is not a product, and it is not owned by any single company. It is a pre-competitive, global protocol designed to ensure that intelligence in tourism works as a system, not as silos.

At its core, Agentic Tourism is about moving from insight to autonomous action - responsibly. Agents can sense context, reason, plan, and act, but always within explicit guardrails defined by humans. Automation handles the expected so that people can focus on what truly matters: empathy, creativity, and trust.

The economic stakes are significant. Tourism already contributes nearly 10 percent of global GDP and is projected to reach $16.5 trillion by 2035. At the same time, AI in tourism is expected to grow to $13.86 billion by 2030. Without shared standards, that growth risks becoming fragmented—captured by closed ecosystems and a handful of dominant platforms—leaving destinations, operators, and communities with limited agency over how AI shapes their future.

At the World Economic Forum in Davos, a closed-door session convened by TOURISE and Globant brought together CEOs, technology leaders, regulators, and policymakers to examine what it would take to introduce the world’s first Agentic Protocol for Tourism. The focus was not on showcasing technology, but on confronting a structural question: how do we govern autonomous coordination across a global, multi-stakeholder system before default rules are imposed by scale or power rather than intent?

The Davos discussion marked an inflection point for Agentic Tourism, not because it introduced a new idea, but because it clarified what execution must look like.

There was clear alignment that coordination will only be adopted if it is proven in the real world. As a result, the initiative is moving decisively from abstract protocol design to live, production-grade pilots, embedded within existing tourism operations and measured against tangible outcomes, such as disruption recovery time, experience continuity, operational efficiency, and traveler confidence.

A second point of consensus was governance. Participants were explicit that trust in agentic systems will not emerge from technical capability alone, but from clear human oversight, explicit guardrails, and enforceable accountability across public and private actors. Governance, in this context, is not an afterthought but a design constraint.

Finally, there was strong agreement on the need for a phased, interoperable roadmap. Rather than forcing a sweeping transformation, the protocol will evolve through bounded execution phases: piloting core agent archetypes, stabilising shared specifications, and expanding participation incrementally. This allows coordination to scale without erasing regulatory, cultural, or operational diversity across destinations.

In short, Agentic Tourism is no longer framed as a future concept, but as an operating hypothesis - one that must now be tested, governed, and refined in the open.

This initiative proposes a practical starting point. Five interoperable agent archetypes—covering experience personalisation, operations optimisation, sustainability, wellbeing, and economic opportunity—are being piloted in real environments, with measurable outcomes. These use cases are deliberately grounded in operational reality, not slideware.

Saudi Arabia’s role in this effort is particularly instructive. The Kingdom is not only investing in scale, but in systems. By convening global actors and anchoring this initiative through TOURISE, it is positioning tourism as a proving ground for how agentic systems can be deployed responsibly at national and international levels.

The larger question is not whether tourism will become agentic—it will. The question is who defines the rules of coordination. If the industry does not shape its own standards, it will inherit them.

(The author is SVP of Solutions, Head of Solution Strategy & Partnerships MENA & APAC at Globant. Any opinions expressed in this article are the author’s own).