The Federal Aviation Administration (FAA) is accelerating long-planned technology upgrades to meet today’s air traffic realities: more flights, new entrants like drones, and rising expectations for resiliency after recent system and workforce impacts. Ongoing efforts to modernize the National Airspace System (NAS) and the Trump administration’s plan to overhaul the nation’s air traffic control system underscore the urgency.
In a recent panel discussion, leaders from General Dynamics Information Technology (GDIT) with deep FAA experience described how modernization is unfolding on two fronts: smarter artificial intelligence (AI)-assisted software practices and cloud-ready platforms designed for mission resilience. As former FAA official and current GDIT transportation leader Mike Hawthorne put it, “We’re not talking about replacing air traffic controllers with AI. We’re not talking about replacing pilots with AI. What we’re talking about is making the systems and the platforms that they use better so that they can do their jobs better.”
Drivers: scale and criticality of mission
The NAS is vast and demanding. As Hawthorne noted, the FAA manages roughly 30 million square miles of airspace and supports about 45,000 flights with 2.9 million people each day, while commercial aviation contributes an estimated 5 percent of U.S. gross domestic product. The scale makes continuous safety and efficiency improvements essential, and it raises the bar for any technology change.
The January 2023 nationwide outage of the Notice to Airmen (NOTAM) system, a pilot messaging database, and subsequent disruptions this year have kept safety and resilience in the spotlight. The Department of Transportation says NOTAM is just one of many legacy systems due for reinvestment, and the Trump administration’s plan, issued in May, outlined a three-year strategy to modernize the entire air traffic control system.
“The FAA’s investments in infrastructure and technology are not just improvements; they are all critical to ensuring the reliability and safety of the aviation industry in the coming decades,” the plan says. “This investment will enhance operational efficiency, reduce flight delays, improve safety outcomes, and support the growing demand in commercial aviation and emerging entrants.”
Software strategy: safer and faster
GDIT’s approach to modernization emphasizes AI-assisted software development across the full lifecycle, from requirements to deployment, with layered testing appropriate for safety-critical systems.
Jay Olsen, who leads GDIT’s Mission Software Center of Excellence, stressed a “human-in-the-loop” posture: “Think of it as the AI sitting on the shoulder of the developer,” while engineers retain control over outputs and releases.
Cloud strategy: a common platform
On infrastructure, Hawthorne emphasized the FAA’s long runway with cloud, which began more than a decade ago in mission support. GDIT is the FAA’s secure gateway to the cloud, offering two dedicated data centers with connectivity to the FAA Wide Area Network. Working with strategic providers Amazon Web Services and Microsoft Azure, GDIT maintains more than 200 critical FAA cloud accounts.
The next step with cloud at the FAA is methodically shifting appropriate NAS capabilities to the cloud in a hybrid model, noted Hawthorne.
Cloud “actually works better than the racks in the basement scenario,” he said, pointing to availability, resiliency, and security gains. He also mentioned momentum behind a long-discussed common automation platform.
The common automation platform is “built on the idea that you shouldn’t have to own and operate … a different set of infrastructure with different types of software and different types of applications for different types of air traffic controllers,” Hawthorne said. “Cloud allows you to have a common automation platform to run the system, everything from traffic flow to arrivals and departures. You’re running on commonly developed and managed applications. There was no feasible way to do that before cloud.”
Application strategy: from monoliths to microservices
Modernization isn’t just about moving compute. It’s about redesigning applications to fit mission outcomes and operational constraints. “We really want to consider re-architecting in a way that makes the most sense for the agency,” Olsen said, adding that hybrid patterns are common, such as keeping sensitive data on premises while containerized applications run in the cloud and using event-driven designs for scalable, stoppable workflows.
Breaking up large, monolithic codebases is a major accelerator of modernization. Some operational automation systems still carry three to four million lines of code, and modernizing is difficult and time-consuming because it’s hard to test and ensure that new changes don’t break old parts of the system, Hawthorne observed.
This is why technologies like containers, virtualization, and microservices are so important, he said. They allow developers to break up a single large system into smaller, manageable pieces, which makes it much faster to update or fix parts of the system without sorting through millions of lines of code.
Next steps: resilience and results
Policy momentum, operational need, and technology capability are aligning. Modernization and workforce initiatives, including the administration’s plan to overhaul the air traffic control system in three years, are responding to urgent needs to strengthen safety and resilience and prepare for the future. AI is an enabler for safer, faster modernization – “a force multiplier in the development world,” said Michael Cole, GDIT’s federal civilian chief technology officer, while cloud becomes the common fabric for consistent services and data governance across the NAS. The goal is a system built for continuous, safe change in a “no-fail” mission.
To learn more, view the discussion and gain additional insights.