The next generation

Travelport’s SVP Product Fahim Khan says future travel will be AI-native, built around intelligent tools, automation and reliable infrastructure—and that future businesses will also be entirely recognizable as an evolution of today’s providers.

This blog series has been fascinating. It’s confirmed the fundamental importance of strong platforms that make AI possible, and explored the many opportunities AI presents, from operational efficiencies to turning inspiration into bookable reality. So what will the next generation of travel look like?

Later in this blog, you’ll hear from a bona fide visionary. When the founder of Booked AI makes statements like “One single AI agent will replace the entire travel agency stack,” we should sit up and take note.

Travel sellers transformed

The future of travel selling will be defined by companies that successfully integrate AI models into the most robust platforms and infrastructure. It will mean using AI in the right ways, at the right time, as a flexible and highly personalized data-driven decision engine.

We’re dealing with two distinct camps, where both sell travel but work in different ways. On the one hand, leisure travel and OTAs are typically non-deterministic. It’s about being highly competitive, fighting for every booking, and trying to build loyalty. The focus is on driving conversion rates—and you’re only as good as your last transaction. AI helps to sharpen open-ended queries and ideas into actionable outcomes, using prompts and personalization to narrow options and improve conversion. When I’m going on holiday with my family (my kids are various ages) we’re interested in doing different things. We need something for everyone. But we want to go together, to a warm location, on these dates. Flexibility and variability are key here, with AI (and human agents) drawing on personal preferences and context where possible.

By contrast, business travel is highly deterministic: I know exactly where I need to be, at what time, and my calendar is planned to within an inch of my life. Here, AI optimizes predefined requirements (e.g. timings, location, corporate policy), delivering precise, compliant and efficient options—again, when powered by strong models using reliable data. Consistency and reliability are critical in this environment. For business travel, your AI models need to provide additional levels of assurance.

The role of infrastructure, data, and orchestration

In any use case, good AI depends on rich user data (history, preferences, behaviors) to generate relevant, ready-to-book options quickly, helping users move faster through the booking funnel. Manual research is replaced by conversational or agentic interactions. That’s where AI will work well for OTAs. This approach combines probabilistic discovery (e.g. suggesting where to go) with deterministic execution (specific flights, prices, itineraries). And it’s already happening. We at Travelport are currently developing a probabilistic search API, utilizing the huge amount of valuable data we have. This will deliver more relevant, context-driven results, rather than going all around the houses first. It will take people down the funnel faster.

Systems must transform broad queries into bookable results through probability-based filtering. APIs and platforms need to simplify and accelerate AI interactions by bundling complex processes into clean and efficient workflows: automated quoting, itinerary generation, disruptions, and rebooking. Providing end-to-end lifecycle support.

But ultimately, the value of an AI-native travel agency will lie in rich validated data— its quality, scale, and structure—and how it’s aggregated, orchestrated, and delivered into the front-end AI models. These are areas where Travelport has invested heavily in recent years. Getting the fundamentals right is essential; trust and data quality are foundational. The strongest platforms provide the best assurance, reliability, and governance to support AI-driven decisions. And users are not yet ready to fully delegate critical decisions to AI—human validation remains important.

Whether AI gets full autonomy or not, it seems likely that future travel agencies will be AI-augmented rather than fully AI-replaced.

AI changes everything but only if you have the right foundations

AI can automate and accelerate most tasks, especially repetitive activities like search, quoting, and administration—we’re currently developing a quoting assistant—and increasingly support complex scenarios such as disruption. But human involvement will remain critical for trust, oversight, and high-stakes decisions.

AI models are advancing rapidly. But data and orchestration remain the real differentiators. The winners will use rich data and intelligent systems to turn vague, non-deterministic requests into high-quality, bookable options—quickly and at scale. This is why Travelport’s partnership with Cognizant and Anthropic is so significant. In a rapidly changing travel distribution landscape, we are redefining the rules: connecting AI tools that can inspire, reason and plan with scalable platforms able to deliver bookable reality.

AI capability is improving daily. One possible future path can be seen in Booked AI. But there will always be a place for people, for that bespoke, human service. Over time, AI will prove to be an enabler: helping us do what we already do—faster, better, and more efficiently.

One thing is certain. AI-native agents will have a headstart over the competition in terms of hyper-personalization, customer service, strategic insights, operational efficiency, and cost optimization. And that’s what competitive advantage looks like.

This is the fifth and final in a blog series exploring how AI is transforming travel, and how the businesses that succeed will be those with the infrastructure, platforms and expertise to make AI work at scale. To catch up with all the previous AI blogs, check out the AI Hub.