FAA tests $12B SMART system
- Transportation Secretary Sean Duffy and FAA officials disclosed new details on SMART in April, an AI planning tool for forecasting airspace congestion days to months ahead. - The key claim is not autonomous control but earlier scheduling decisions — with vendors testing models that could spot bottlenecks hours before flights stack up. - It matters because SMART sits inside a much larger ATC rebuild, where officials say $12.5 billion is only a down payment.
Air traffic control is getting an AI layer — but not the kind that lands planes for people. The FAA’s SMART system is a planning tool, basically a giant prediction engine meant to see congestion, weather trouble, and routing conflicts before they turn into delay cascades. That matters because the U.S. system still runs on a lot of old hardware and fragmented software, and every bad weather day exposes the seams. What changed in April is that Transportation Secretary Sean Duffy, FAA Administrator Bryan Bedford, and industry executives started publicly filling in what SMART is supposed to do — and what it is not. ### What is SMART, exactly? SMART stands for Strategic Management of Airspace Routing Trajectories. The idea is to model the national airspace far enough ahead that airlines and traffic managers can reshape schedules before the system jams up. Think less “robot controller” and more “forecasting dashboard for the whole network.” The FAA’s broader pitch is that smarter planning should cut delays, cancellations, and last-minute traffic restrictions. (flyingmag.com) ### Is this replacing controllers? No — and FAA officials and vendors have been unusually explicit about that. SMART is not meant to do safety-critical separation, the real-time work controllers do to keep aircraft apart. Human controllers still own that job. SMART sits upstream from them, helping decide how to spread demand across airspace and time so controllers are not handed a mess in the first place. (flyingmag.com) ### What problem is it trying to solve? Today’s system is still too reactive. Controllers and traffic managers can handle immediate conflicts, but the whole network often gets managed in short windows, after weather shifts or demand spikes have already started to pile up. SMART aims to move that horizon outward — from near-term tactical fixes to planning days, weeks, or even months ahead, depending on the use case. That is the real promise here: fewer domino-effect delays because the dominoes get rearranged earlier. (flyingmag.com) ### How far ahead can it see? This is where the reporting gets messy, because different descriptions refer to different horizons. One version of the pitch is strategic forecasting months ahead for traffic and weather patterns. Another is much more concrete — spotting sector overloads or flight conflicts roughly two hours before they become acute. Those are not contradictions. They are two layers of the same idea: long-range planning plus shorter-range deconfliction before wheels-up. (flyingmag.com) ### Who is building it? The FAA has brought in Palantir, Thales, and Air Space Intelligence to compete on the tool. That does not mean one of them has “won” SMART outright yet. It means the agency is testing competing approaches while it pushes a much bigger modernization program that also includes new communications, surveillance, and automation infrastructure. (flyingmag.com) ### Where does the $12 billion number fit? The catch is that $12 billion is not really the price tag for SMART alone. The official number the administration talks about is a $12.5 billion initial funding tranche for a brand-new air traffic control rebuild, with FAA leaders saying they will need about $20 billion more to finish the overhaul. SMART is one piece inside that larger project — alongside the Common Automation Platform, telecom upgrades, radars, and facility work. (bloomberg.com) ### So what should travelers believe? Do not expect an AI miracle by Memorial Day. SMART sounds useful, and the logic is solid — predict bottlenecks earlier, make fewer desperate fixes later. But the hard part is not the demo. It is integrating forecasts into airline schedules, controller workflows, and a national system the FAA is trying to rebuild by the end of 2028. That is a huge lift. ### Bottom line SMART is real, but the viral version of the story overshoots. (transportation.gov) This is not an AI autopilot for the skies. It is a still-developing planning tool inside a far larger FAA modernization push — promising, expensive, and only as good as the system around it. (flyingmag.com) (faa.gov)
