Artificial intelligence (AI)I is revolutionizing the aviation industry by simplifying processes that were previously completed manually by humans. AI uses mathematical and data analysis processes to mimic human decision-making and has the potential to transform airport operations (Strome, 2024). Subcategories of AI include machine learning (ML) and deep learning (DL) which involve algorithms that model human decision-making processes and learn from available data inputs. Generative AI is the most well-known type of AI that can create written content, images, or videos using DL processes (Strome, 2024). 

Central to the modernization of the national airspace system (NAS), is the integration of AI technologies into the airport environment. There are a variety of benefits of AI use for airports including facilitating a smoother passenger travel experience, enhanced accessibility, strengthened security, operational efficiency, optimal resource allocation, and data-driven decision making (Crist, 2024). Airports are now developing specialized research and development (R&D) innovation teams to identify and implement new AI technologies at their airports. Their goal is to identify data needs, potential AI applications, and develop implementation plans for the airport.

However, Crist (2024) explains that there are some unique challenges presenting themselves as airports begin to implement AI. First, airports tend to implement too many AI technologies at once, instead of a few well-done focused applications. This can lead to stalled progress caused by resource limitations, including human and financial hurdles due to resources being stretched too thin. Secondly, there is a steep learning curve for technical teams and business partners which makes it difficult to implement AI. Executing multiple AI technologies at once results in an even steeper learning curve for R&D teams. Finally, Crist (2024) found that AI requires an accurate, timely, and robust data source to be effective. Data must be collected, cleaned, centralized, and organized before it can by analyzed by AI technology. In many cases, data belongs to business units across the airport including airport management, airlines, air traffic control, concessionaires, and other airport tenants. Data must be gathered from these airport stakeholders before it is useable for AI analysis, which is difficult and labor intensive.

The key to effective AI realization is the establishment of a foundational data plan with proper data governance, infrastructure, analysis, and innovation (Crist, 2024). Many airports have created a Chief Data Officer (CDO) position that is responsible for leading the organization’s data management processes. The CDO is responsible for developing the airport’s data strategy and aligning AI initiatives with the airport’s business goals. Also, the CDO leads the development of proper data governance and establishes policies for data security and accessibility. Data infrastructure must be developed to store the airport’s data and make it accessible to airport data analysts to ensure a seamless AI implementation. These processes are part of the overarching data governance framework that is defined by the CDO and applies to airport stakeholders as they collect, clean, and organize business data.

Furthermore, Strome (2024) explains that the reliable use of AI technology is highly dependent on the availability, quality, and the governance of data. To embrace the full potential of AI technology, airports must ensure that they have developed and implemented sound data governance practices. Otherwise, the dataset used by AI will be incomplete which leads to inaccurate data outputs and decision-making. By establishing clear rules of engagement regarding the collection, analysis, and use of data, airports will be prepared to implement AI technologies that produce accurate and useful results (Strome, 2024).

Further research highlights the importance that data used in large language models (LLM) be managed responsibly, confidentially, securely, and ethically (Saurabh et al., 2024). According to Saurabh et al. (2024), there are five things that must be in place for LLMs to be responsibly governed. These include ethical data stewardship, privacy, security, performance and risk monitoring, and regulatory compliance. The authors argue that unregulated data practice flows and inadequate governance frameworks across various industries create risk factors that can be mitigated by effective governance systems. This can be seen in the aviation industry where data is siloed between different airport stakeholders including airport management, airlines, security, concessions, and other airport tenants. Each organization has its own data governance processes that leads to fractured data collection, analysis and difficulty applying AI technologies across multiple airport stakeholders.

Therefore, further research is needed to develop scalable AI governance methods that are unique to the airport environment and can accommodate a variety of data sources from airport stakeholders (Saurabh et al., 2024). With the added layers of differing airport management structures, airport classifications, government regulations and operational characteristics, airports need data governance methods that consider the dynamic environment of airport operations. This is the first step in standardizing AI implementation across airports and can lead to scalable AI integration throughout the NAS.

References

Crist, C. (2024). Don’t be left at the gate: A practical guide to AI adoption for airports.Journal of Airport Management, 19(1), 6–14. https://doi.org/10.69554/NWFQ3298

Pahune, S., Akhtar, Z., Mandapati, V., & Siddique, K. (2025). The importance of AI data governance in large language models.Big Data and Cognitive Computing, 9(6), 147. https://doi.org/10.3390/bdcc9060147

Strome, T. (2024). Data governance best practices for the AI-ready airport.Journal of Airport Management, 19(1), 57–70. https://doi.org/10.69554/ZHBT2608