What a year for AI. Progress moves so fast that we now measure breakthroughs in months, not years. Aviation, by contrast, evolves cautiously, not because of a lack of innovation, but because adopting new technology safely, at scale, is fundamentally hard. That tension between rapid AI evolution and deliberate aviation adoption is now impossible to ignore, and it will define how artificial intelligence actually enters aviation operations in 2026. 

In the AI domain, at the beginning of last year, we had not yet discovered the Deep Research capabilities in ChatGPT, we had not seen new reasoning models (such as DeepSeek, Gemini, or Llama), and Agentic-AI technology was only just beginning to enter the conversation. 

By contrast, in the aviation sector, much of the discussion around artificial intelligence over the past year was largely exploratory: pilots, proofs of concept, and experiments. Many organizations began to internally discuss how they could start using these tools realistically and effectively within their businesses and, perhaps more importantly, what risks their adoption could introduce, given a safety-critical environment. However, the results were spotty.

Despite the technological momentum, the aviation sector does not adopt technology at the same speed at which it is developed. This is the result of a long history of digitalization. Operations, for example, involve multiple stakeholders, countries, and regulatory environments; in other words, a small change scales in complexity, benefits, and risks. The challenge is not innovation, this is very much like aviation itself. Rather, the challenge lies in the reliable and safe adoption of new capabilities that do not compromise safety or operational efficiency. 

This does not mean that aviation lacks an appetite for AI. On the contrary, that appetite clearly exists, but it is in tension with the need for the adoption of this new technology to be consistent with the sector’s legacy, its operational environment, and, above all, to justify the return on investment. Coming out of the World Economic Forum in Davos, the mood around AI has become considerably more serious: we are now looking at return on investment, system resilience, and the degree to which people trust these systems. 

With this in mind, ALG’s Digital & AI expert team has engaged with multiple Data & AI experts (researchers, CDOs, and industry leaders from ALG and its clients) to develop a perspective on what we can expect in the year ahead. This blog post does not aim to provide an exhaustive view of all AI applications across the aviation sector. AI is already being applied in many areas beyond those covered here, and its scope will continue to expand rapidly. Instead, the trends presented are selected, emerging technological directions that are expected to become particularly relevant in 2026, with the main focus on operations.  

Also, these trends should not be understood as a large-scale revolution of our sector, but rather as signals and sources of inspiration that point to the early, tangible examples that will begin to emerge as AI adoption progresses. The following sections explore how this transformation is unfolding for airports, airlines, passengers, air navigation service providers, and regulatory authorities. 

Can airports “think”? 

More intelligent operations with Agentic AI 

Agent-based systems revitalized large language models in 2025, moving them beyond purely conversational tools toward more autonomous systems. While this evolution is still at an early stage, its potential for the aviation sector in 2026 and beyond is significant. AI agents are autonomous systems designed to achieve high-level objectives, interacting with other systems and tools (AI-driven or not) and adapting to new situations with minimal human supervision. 

Consider the failure of a shuttle transporting passengers from the aircraft to the terminal. Today, assigning a replacement shuttle typically requires several manual communications, introducing delays of five to ten minutes. In a monitored and automated environment (for example, with shuttle geolocation) AI-based agents could immediately identify and dispatch the optimal available vehicle, reducing response time to under a minute and preventing the disruption from propagating. 

In 2026, we expect to see more initiatives aimed at evolving the airport operational concept, expanding the TAM (Total Airport Management) concept. Multimodal AI will interpret the Airport like humans and Agent-to-agent communication might start to pop-up in conceptual conversations.

Now imagine this concept at scale. Under this model, AI moves from being a set of isolated use cases to becoming part of the operational fabric of the airport. AI agents applied to operations can link information from multiple systems, detect deviations early, and act proactively, helping to anticipate incidents before passengers feel the impact. When these agents are continuously embedded into infrastructure and processes, they give rise to what is often referred to as ambient intelligence: an invisible layer that perceives the state of the airport, reasons over it, and triggers coordinated actions, often without passengers even realizing that an AI system is involved. 

In this line, platforms such as iTAM, which will be developed by Abu Dhabi Airports and SITA, aim to become an example of how new technologies can contribute to optimizing airport operations. The platform proposes the real-time integration of information from airlines, ground equipment, and air traffic control, applying artificial intelligence models to anticipate incidents before they occur. From a technological point of view, this type of solution relies on streaming data architectures and predictive models, such as recurrent neural networks or time series models, capable of detecting patterns and deviations in advance. Complementarily, large language models, such as those developed by OpenAI (the creators of ChatGPT), are being incorporated, allowing operations managers to consult information in natural language with traceability and reliability.  

From there, Generative Business Intelligence (GenBI) could take the center stage, allowing managers to work with data in a much more direct way, asking questions about bottlenecks, risks, or alternative scenarios, and obtaining answers accompanied by explanations based on operational context.  

Technology as the passenger’s best friend 

Digital identity, biometrics, and AI-based personalization 

In 2026, the passenger experience will continue to be one of the main areas of innovation in the sector. Airports Council International (ACI) World reveals key trends shaping air travel in 2026 with the launch of its latest Airport Service Quality (ASQ) Global Traveller Survey. The survey, conducted with 4,125 passengers across 30 countries, indicates an increase in demand for human-centred, personalized, and memorable airport experiences. 

There is growing interest among passengers in solutions that facilitate their journey through airports, ranging from remote check-in and off-airport bag drop to biometric verification. In fact, 72% of surveyed passengers stated they would be willing to use biometric solutions. As a result, in 2026, we expect multiple airports to continue deploying identification systems at key points along the journey, such as security checks, border control, and boarding. 

There are multiple initiatives driven by international organizations that aim to support this adoption. For example, IATA One ID seeks to give passengers control of their digital identity throughout their journey, from check-in to boarding, using facial recognition and digital credentials instead of physical documents. Similarly, ICAO promotes the use of digital passports (Digital Travel Credentials), enabling passengers to identify themselves without a physical document. This change reflects a shift towards identification systems integrated into airport processes, in which identity verification is carried out continuously and automatically. The use of a unique biometric identifier, managed securely, reduces the need for repetitive checks, while edge computing allows identity to be validated in real time at checkpoints, without total dependence on remote cloud services, increasing the efficiency and resilience of the system. 

Everything points to travel becoming a simpler experience in 2026, adapted to each individual and with fewer visible steps for passengers. Digital identity speeds up checks, mobile apps offer personalized services, and artificial intelligence works in the background to make everything run more smoothly, with less waiting and more convenience. This will allow passengers to focus on enjoying their trip, without worrying so much about the paperwork. 

On the other hand, the ACI survey shows that fewer than 50% of passengers believe that airport staff demonstrate empathy and proactive support throughout their journey. This highlights an opportunity for improvement in the creation of personalized solutions that can represent a clear before-and-after in the passenger journey. For 2026, we expect that, with the increasing availability of AI tools, multiple solutions focused on enabling personalized passenger support will begin to flourish. 

Mobile applications may allow passengers to pre-book duty-free purchases, restaurant tables, or access personalized services within the airport. A good example is Singapore Changi Airport, where digital services combine real-time operational data with passenger applications to provide immediate benefits when disruptions occur, such as refunds, access to VIP lounges, or flight rebooking options—all activated automatically. 

Why now? From a development perspective, it could be expected that the rise of low-code and no-code platforms might accelerate the creation of these digital experiences. Airports and airlines will be able to prototype and launch specialized applications much more quickly, testing new services without large upfront investments. In addition, if we couple this approach with technologies such as edge computing, these systems can be deployed in environments with different security requirements, for example, at security checkpoints or boarding gates. 

The new logic of airlines 

LLMs, AI agents, and automation for decisions on tight margins 

While AI already supports a wide range of airline activities, from maintenance and fuel optimization to revenue management and customer service, its role in operational decision-making is beginning to expand. One emerging trend for 2026 is the use of large language models and AI agents as integrators of decision-making in highly volatile operational environments, where speed, consistency, and resilience are critical. As IATA has repeatedly highlighted, operational resilience and the ability to respond quickly to unexpected events are increasingly what separate airlines that protect value from those that struggle to absorb disruption.

What is starting to change is the role artificial intelligence plays in operations. By 2026, the differentiator will not be whether airlines use AI, but how effectively it is embedded into decision-making under pressure. The shift is less about deploying isolated use cases and more about integrating intelligence into operational workflows, supporting faster and more consistent responses to disruptions, crew and aircraft constraints, and rapidly changing market conditions. 

In 2026, the competitive edge for airlines will not come from better forecasts but from faster and more consistent decisions when things go wrong. AI can become a differentiator when it helps protect revenue, reduce disruption costs and stabilize operations under pressure. 

Airlines that treat AI as an integrator and orchestrator—rather than as a collection of disconnected tools—are better positioned to navigate volatility, protect margins, and build more resilient operations. To enable this shift, large airlines are investing in consolidated data platforms that act as a foundation for AI-driven decision support. Initiatives such as the data hubs developed by easyJet or IAG illustrate a move toward platform-based models, where operational, commercial, and customer data are brought together to support automation and coordination across the organization.

This same logic underpins initiatives like Nevio, the modular offer and order management platform being developed by the Lufthansa Group with Amadeus. Beyond dynamic pricing or personalized offers, these solutions point to a deeper change: the convergence of commercial and operational decision-making, supported by AI models trained on historical demand, customer behavior, and real-time operational constraints (fleet availability, slots, crew). In this context, large language models are beginning to act as natural interfaces to complex data environments, providing explainable and traceable insights rather than opaque recommendations. 

However, capturing this value at scale remains challenging. AI adoption in airlines is constrained by the need for strong data foundations, clear governance, and robust human oversight in safety- and revenue-critical processes. As a result, 2026 is unlikely to mark a sudden breakthrough, but rather a period in which early, well-prioritized initiatives begin to demonstrate tangible return on investment. 

Rethinking airspace management 

ATM platforms and the future of Digital Twins 

Technologically, airspace management is undergoing a gradual but profound transformation. Traditionally, air traffic management (ATM) systems were built on closed, monolithic architectures developed by technology vendors using proprietary hardware. The evolution of these systems was complex, particularly at their interaction points, which made upgrades slow and costly. The current focus is on evolving these systems towards more open, modular, and flexible digital architectures, driven by initiatives such as the SESAR ATM Master Plan, developed under the framework of the SESAR Joint Undertaking.

This evolution is commonly described under the concept of Platform as a Service (PaaS). In the ATM context, this refers to the provision of common digital platforms that decouple operational services from legacy hardware layers. These platforms enable ATM functions to be implemented as modular services, built on common digital platforms shared across providers and deployed on flexible infrastructures that allow ATM services to evolve independently while remaining aligned with the safety-critical nature of the environment. 

For example, in Norway, Avinor and Indra have launched a center that controls up to 23 regional airports from a single point. And in the European network, Eurocontrol already uses cloud platforms as part of the iNM program, allowing data and decisions to be managed from anywhere, without relying on fixed physical architecture.

New ATM architectures will set the right foundations for AI adoption, at scale. ATM might create the right environment for AI to shift from giant models to domain-specific reasoning systems tailored to aviation safety-critical operations. 

This infrastructure will act as an enabler for AI-based use cases. Many of these are not new by 2026; for example, systems capable of predicting traffic conflicts, optimizing routes, and anticipating adverse conditions are already in use, helping controllers make faster and safer decisions in each operational situation. 

We also expect to see further progress in the Airspace Digital Twins roadmap. These digital twins are virtual replicas of the real operational environment that combine real-time operational data, simulation models, and predictive algorithms. While the concept itself is not new, it will be particularly interesting to observe how the integration of AI agent–based architectures begins to mature, operating within these digital twins, testing thousands of possible scenarios, and helping to assess the potential outcomes of different operational decisions. This approach allows changes to be tested and validated without putting the actual operational environment at risk, making the ATM system safer, more efficient, and easier to adapt.

Civil Aviation Authorities and Agencies 

LLMs, automation, and the challenge of certification and assurance 

As aviation accelerates its technological transformation in operations, maintenance, and traffic management, aviation authorities are also incorporating AI into their regulatory agenda. In 2025, the European Union Aviation Safety Agency (EASA) opened its first public consultation on Artificial Intelligence in aviation with the publication of the Notice of Proposed Amendment (NPA 2025-07), which proposes an ‘AI trustworthiness’ framework aligned with the EU AI Act, including assurances requirements, human factors, and ethics for AI based systems, and prepares the industry for future requirements for automated assistance and human-AI teaming.

AI resilience will become critical. 2026 could be a crucial year in terms of how aviation regulations incorporate technologies that are already revolutionizing the industry. EASA is leading a structured approach to AI assessment, certification, and assurance, while other agencies such as the FAA are developing their adaptive strategies.  

Moreover, EASA has been working for years on its AI roadmap, which identified different levels of applications (from basic assistance to close collaboration between humans and systems) and plans to include advanced techniques, including those based on generative models, such as LLMs, in future regulatory work.

The focus on assurance and certification is crucial because traditional aeronautical certification methods were not designed for technologies that learn from data and make decisions based on complex models. Therefore, the NPA and EASA activities seek to define how to establish trust, traceability, bias mitigation, and comprehensive testing AI systems applied to critical functions, preparing the ground for uses such as automation of regulatory or supervisory tasks based on LLMs to be aligned with high safety standards. A good example of this approach is the Federal Aviation Administration (FAA) in the United States, which has published its Roadmap for Artificial Intelligence Safety Assurance. In this document, the FAA defines the basic guidelines for adapting certification and safety processes to AI-based systems. The roadmap emphasizes working together with industry and other regulators, preparing teams for these new challenges, and exploring how AI itself can help analyze large volumes of information related to certification and regulatory compliance.

In this context, the capabilities of LLMs are beginning to be considered not only for operational tasks but also to support regulatory activities, such as automating compliance reviews, generating certification artifacts, or even assisting in complex risk analysis. However, authorities emphasize that these applications must be accompanied by robust assurance frameworks that allow us to understand, verify, and validate their behavior before formally incorporating them into critical certification and oversight processes. 

Conclusion

The AI trends shaping aviation in 2026 do not evolve in isolation. Regulation, platforms, operations, passenger experience, and airport environments are advancing at different speeds and, most probably, the value captured from AI will ultimately be defined by the weakest link in that chain. 

The question for aviation is not whether AI will take off. Reflections and opinions from key experts and world leaders converge on the fact that it will. The real question is whether organizations will be ready to adopt AI solutions as a strategic, system-wide capability, delivering value without compromising resilience, safety, or trust.  

After initial years of pilots and experimentation, the industry is entering a phase where isolated successes will no longer be enough. However, this real return on AI investments will only emerge if AI initiatives are prioritized and aligned with the operational and regulatory realities of the sector. In other words, ROI will come as far as AI is conceived as a technological layer supporting a clear CONOPS with measured benefit. AI technology (in its many forms, not only limited to LLMs) will continue to provide value to aviation in 2026. From ALG, we are confident that these benefits are no longer theoretical: for the right players, they are about to take off. 

About the authors

Antonio Cabeza

Aeronautical engineer with a MSc in Big Data and Advanced analytics. Specialises in strategical projects in airport and air traffic operations with wide expertise in digitalisation and regulatory related projects. Senior Engagement Manager at ALG.

Francisco Serrano

Aeronautical Engineer with an MSc in Air Transport Systems, specialized in air traffic management and data-driven operational performance. Experienced in applying advanced analytics and machine learning across complex aviation and air navigation environments. Consultant at the AI and Digital Transformation at ALG.

Leticia Fernández Abella

Aeronautical Engineer with a Master’s degree in Deep Learning for audio and video signal processing and Big Data. Experienced in data analysis in aviation safety. Business Analyst in the Data team at ALG.