AivexaNewsSearch
AI news for builders and product teamsChecked every hour

FAA expands AI advice to manage air traffic congestion across entire US airspace

Collected Nov 12, 2026

The Federal Aviation Administration has widened the reach of an artificial intelligence tool that advises air traffic controllers, moving it from a regional pilot to the airspace handled by all 22 air route traffic control centers in the United States. The agency announced the expansion in a statement dated November 10, describing it as a follow-on to an initial deployment in the National Capital Region in September. The system, called SMART, has cost $875 million, according to the FAA.

The agency's numbers from the first phase are the main argument for going national. Since going live over Washington, DC, on September 21, SMART has helped reroute 180 flights. The FAA says that produced an average saving of about 10 minutes and nearly 80 gallons of fuel per flight. The agency's stated goals for the wider rollout are fewer delays, lower fuel burn and a smoother experience for passengers.

The FAA describes SMART as a cloud-based system that pulls together weather patterns, flight paths, traffic flow and controller staffing metrics to forecast how traffic will develop. From those inputs it generates scheduling recommendations, suggests more efficient routes and works out the best sequence for arriving aircraft. In effect, it is a planning aid that tries to smooth congestion before it turns into holding patterns and ground stops, rather than a system that issues instructions to pilots.

The human chain of command is explicit. A team of FAA specialists in Washington reviews and validates SMART's recommendations before they reach air traffic control facilities around the country. Controllers at the local level can accept or reject what they receive. The FAA said in its September 21 announcement that SMART cannot replace controllers or take control of an aircraft. This suggests the agency is positioning the tool as decision support, with accountability staying with people, which is the usual pattern for AI in safety-critical operations.

Support and concern have arrived together. Delta CEO Ed Bastian and American Airlines CEO Robert Isom both offered supportive statements when the initial deployment was announced, and airline executives have backed the national expansion. The National Air Traffic Controllers Association, which represents more than 15,000 members, told Flying magazine that it was not consulted or involved in the design, testing or implementation of SMART, and said it therefore cannot speak to how the system affects the controller workforce. That gap is a notable one: the people expected to act on the system's output had no formal role in shaping it, at least according to the union.

The rollout has proceeded in steps. After the DC-area launch, coverage was extended to the air route traffic control centers serving Atlanta and Jacksonville and Miami in Florida, according to The Dallas Morning News. The FAA gave that newspaper an example in which SMART recommended a shorter, faster route for a cargo flight bound for Dallas-Fort Worth International Airport; the flight was not identified, and the agency has not offered detail on how the system has assisted passenger traffic specifically.

For a developer or product reader, the interesting design questions are the ones the public record does not yet answer in technical depth: how the system weighs staffing metrics against weather and traffic, how recommendations are surfaced to controllers, and what feedback loop exists when a suggestion is declined. In practice, an advisory tool of this kind lives or dies on acceptance rates. If controllers frequently override SMART, the projected fuel and time savings shrink; if they follow it closely, the quality of its forecasts becomes operationally important. A likely trade-off is that national-scale deployment broadens the data pool and could improve predictions, while also exposing the system to a much wider range of traffic patterns than the DC region offered.

The airline industry's enthusiasm is easy to understand, since time and fuel savings translate directly into cost and schedule reliability. The union's distance from the program is the counterweight, and it points to a familiar adoption problem: tools that change how frontline staff work tend to land better when those staff help test them.

Why it matters: Controllers across the country, airline operations teams and, by extension, passengers are now inside the footprint of a federally deployed AI advisory system with a large price tag and a short track record. For product teams building decision-support tools in regulated, safety-critical domains, the SMART model—machine recommendations filtered through a central human validation team and optional at the point of use—is a template worth studying. How often controllers accept its advice, which the FAA has not disclosed, will be the number that determines whether the projected savings hold at national scale; that is my inference, not a stated agency figure.

Read at Ars Technica · AI

Based on reporting from the original publisher. Visit the source for full context and later updates.

Publisher excerpt

Airlines praised the AI tool, but the air traffic controllers’ union wasn’t involved.