Warum die Reisebranche eine gestufte KI‑Einführung benötigt, nicht ein Rennen zur Autonomie

KI‑Diskussionen in der Reisebranche drehen sich häufig um die Frage, wie die Technologie den Sektor schnell transformieren und hochwirksame Ergebnisse liefern kann. Die Realität ist jedoch differenzierter. Während KI in Bereichen wie Personalisierung, Kundenservice und operativer Effizienz greifbaren Nutzen gezeigt hat, übernehmen nicht alle Akteure sie im gleichen Tempo. Ein wesentlicher Grund dafür ist das komplexe technologische Ökosystem.

Fluggesellschaften, Online‑Reisebüros (OTAs) und Travel‑Management‑Companies (TMCs) agieren in über Jahrzehnte aufgebauten, miteinander vernetzten Netzwerken. Viele setzen weiterhin auf fragmentierte Datenumgebungen und Altsysteme, die die Geschwindigkeit der KI‑Einführung und -Skalierung einschränken. Verstärkte Bedenken hinsichtlich Transparenz, Verantwortlichkeit und Zuverlässigkeit verschärfen das Problem zusätzlich.

While most travel companies talk about AI as a single transformation story, its adoption is unfolding across three distinct yet interconnected layers: (i) progress in customer-facing and operational automation, (ii) friction created by legacy infrastructure, and (iii) the lack of institutional trust. Each is advancing at a different pace, creating unique dynamics that demand tailored responses. Understanding these differences provides a more practical framework for identifying where the greatest opportunities (and risks) are likely to emerge.

1. Fortschrittsbewertung

Whether searching for flights, managing itineraries, or resolving disruptions, travelers increasingly expect every interaction to be seamless, personalized, and responsive. Gen AI assistants are helping by streamlining trip planning and customer support. In 2025, almost 40% of US travelers used Gen AI to plan trips. At the same time, machine learning models are enabling hyper-personalized offers based on traveler behavior, loyalty preferences, and purchasing history.

AI is also creating immense value behind the scenes. Travel companies are using advanced analytics to better predict demand, manage capacity, strengthen workforce planning, and handle disruptions. For example, a major travel services provider lowered its cost-per-booking by 10% y-o-y with Gen AI, while a Canadian airline reported a 2% uplift in unit revenue and a 10% boost in network-driven revenue through AI-enabled pricing.

Yet, despite growing evidence of the tangible value AI can deliver, legacy systems remain a significant obstacle to widespread adoption.

2. Friktionen angehen

The industry’s legacy infrastructure was not designed to support the real-time, unified data pipelines that AI requires. At the center of the problem is the Global Distribution System (GDS). GDS platforms were architected several decades ago on EDIFACT messaging protocols and still account for the dominant share of indirect airline sales globally. Integrating New Distribution Capability (NDC) with a legacy Passenger Service System (PSS) can take months of testing and development, particularly for airlines offering multiple fare brands or ancillary products. The challenge spans contractual restrictions on content distribution, organizational readiness, and the absence of standardized data across regional markets.

The lack of end-to-end data visibility further constrains an organization’s ability to scale AI effectively.  A GBTA survey revealed that only 12% of corporate travel buyers have a consolidated view of their program data, a foundational constraint that limits what any AI system can deliver, regardless of model sophistication.

Airlines, OTAs, and TMCs are navigating this by adopting hybrid strategies that layer intelligence onto existing systems using agentic-ready APIs. Meanwhile, NDC-native players are taking a different approach, building AI-driven servicing and policy compliance directly into their architecture and reducing reliance on traditional GDS channels.

But technological integration alone does not guarantee success. As AI becomes more deeply embedded and begins to influence higher-stakes decisions, the next challenge emerges: trust.

3. Vertrauen vorantreiben

Research from GBTA shows that while 92% of travel buyers are interested in AI-driven spend forecasting and 89% in automated disruption management, only 57% of the same buyers are comfortable with AI autonomously changing or canceling bookings. This contrast highlights a fundamental trust gap.

Travel leaders are looking for explainable, auditable solutions. Meanwhile, customers increasingly demand transparency and accountability. Building trust across diverse stakeholder groups requires a commitment to ‘responsible evolution,’ balancing innovation with transparency and governance.

This means the organizations making the greatest progress are not necessarily those deploying the most advanced models or moving the fastest. Rather, they are the ones following a structured approach: strengthening data foundations before scaling AI, validating use cases in controlled, high-impact environments, and embedding proportional governance into the transformation journey. Unlike uniform governance, proportional governance organizes AI agents into levels of autonomy, each with defined trust boundaries and governance requirements.

KI‑Ambitionen in nachhaltige Wirkung umsetzen

For travel organizations moving toward the next phase, embracing this sequential approach is no longer optional; it is what separates leaders from laggards. In this emerging context, business process management partners play a critical enabling role by helping organizations operationalize AI across complex ecosystems of data, processes, and human decision-making. Ultimately, organizations that translate disciplined execution into a durable, hard-to-replicable advantage will emerge as winners.