In the many conversations I have had with C-suite leaders across the airline industry, one theme consistently stands out: Expectations, attitudes, and concerns around AI vary significantly from one leader to the next. These conversations also tell me that while technology is moving at real speed, everyone seems to agree that it is hard to keep pace with that change.
Research from KPMG shows that large companies in the region expect to spend an average of US$245 million over the next 12 months, well above the global average of US$186 million. Confidence is also outpacing other regions, with four in ten Asia-Pacific companies expecting AI agents to take the lead on managing specific projects over the next two to three years, compared with 30% globally.
For airlines, agentic AI could change everything from end-to-end trip planning to customer service. But the aviation sector is not a simple testbed. It’s highly regulated, operationally complex, and built on trust. Given this context, how can airlines deploy this technology safely, responsibly, and at scale?
From these conversations, five practical recommendations emerged that I believe can help airlines scale their use of agentic AI more confidently and responsibly.
1. Consider the workflows that deliver quick and measurable impact
One approach for the industry will be to prioritise a handful of workflows where AI is expected to deliver measurable returns. An obvious starting point is call centre automation, where AI voice agents can now undertake predictable interactions such as changing the return flight on an existing booking, reducing wait times, and freeing human teams for more complex interactions.
This allows airlines to test the solution, help build confidence, strengthen internal capability, and create a more credible foundation for scaling AI in more complex parts of the business.
2. Establish the data foundations needed for value creation
As with all data analysis technologies, the value gained from agentic AI will be proportional to the quality of the data it can access. This means airlines need to ensure the data and business logic available within their own systems are well structured, accessible, and ready for AI agents to use effectively.
Specifically, event data generated by an airline’s operations needs to be captured and made available in a data lake to power real-time inference. This also requires APIs to be re-engineered for AI agents, and generic AI models must be fine-tuned to understand the industry’s language.
3. Assess integration readiness and scale in phases
Airlines would benefit from mapping out the applications, data sources, and operational dependencies that agentic AI will rely on. This means assessing whether data is complete and available at the speed required, as well as whether systems can connect securely through reliable interfaces. Airlines can take a phased approach, with initial deployments focusing on lower-risk, human-assisted workflows and, once performance and governance controls are in place, progressively expanding agent access.
This transformation is also taking place at the same time as the modernisation of airline retailing, where a more connected data model can provide a clearer view of the offers available to travellers, manage bookings and orders more seamlessly, and bring customer information together across the journey. Together, these changes will create the data foundation needed for AI to deliver more personalised and relevant experiences for the end user.
4. Build in governance by design
Unlike prior generations of enterprise IT, AI agents are probabilistic and not deterministic. While this makes it possible for them to problem-solve through reasoning, it also means they can be unpredictable and prone to hallucinations and errors. The task for airlines is to harness this new capability in a way that limits risk, which is where governance comes in.
Governance cannot be an afterthought. It is important to consider technical, ethical, legal, and process measures for AI governance from the outset. As AI regulation evolves globally, organisations should strengthen governance through dedicated AI oversight, ethics frameworks, secure technical infrastructure, and compliance programs aligned with emerging standards. Striking the right balance between tech-enabled governance that draws on deterministic, rule-based systems as guardrails for AI agents and human-in-the-loop oversight for each specific use case is critical for risk mitigation.
5. Start change management from day one
AI adoption is as much an organisational transformation for the airline industry as it is a technological one. Without clear communication that explains the change, why it’s happening, and how it can benefit employees and customers, even well-designed systems may struggle to gain trust and adoption.
While AI is poised to automate transactional tasks, humans will continue to play a fundamental role. Airlines have an opportunity to redeploy human experts to new positions focused on creative thought, AI orchestration, and customer service. When the basics are automated, we can focus on more complex scenarios that can improve the traveller experience.
I believe the opportunity for airlines is not simply to automate workflows, but to redesign them, and the urgency to adopt agentic AI will only continue to grow. The International Air Transport Association forecasts that the Asia-Pacific region will grow to 4.1 billion passengers by 2044, an additional 2.4 billion journeys from 2024 moving through airports and airline systems. Meeting this demand will require airlines to become more intelligent, responsive, and scalable without compromising the safety, service, and trust on which the industry depends.
As airlines gradually look to automate more parts of the journey, the opportunity will be having more time to focus on the moments that matter most to travellers. The airlines that can execute on these points won’t simply keep pace. They will gain a competitive advantage in creating new experiences for the agentic age.
