Air travel in America grinds through chronic delays. Controllers juggle flights on outdated systems. Now the Federal Aviation Administration pushes AI into the mix. Transportation Secretary Sean Duffy champions the charge. The tool? SMART, or Strategic Management of Airspace Routing Trajectories. It promises to spot conflicts hours ahead. But skeptics see risks in handing predictions to algorithms prone to errors.
Duffy laid it out bluntly during a CBS News interview. “This software will say, ‘well, listen, we can see this 45 days out. Let’s move some of those flights a little bit later, or five, seven, 10 minutes earlier, and we can resolve the issue. And so then you are not delayed,’” he said. The price tag hits $12 billion initially, part of a broader overhaul. Congress approved $12.5 billion, yet FAA leaders say they need $20 billion more to finish the job, as reported by Bloomberg.
Three firms vie for the contract: Palantir Technologies, Thales SA, and Air Space Intelligence. Palantir confirmed its role in a statement to investors, noting it provides a data analytics tool to advance FAA aviation safety goals, according to Futurism. A person familiar with the matter told Bloomberg the FAA tapped these players to compete. The Air Current detailed how SMART shifts control from reactive to predictive, with Duffy explaining at a Semafor event: controllers could tweak paths 1.5 to two hours before clashes, not mere minutes prior, per The Air Current.
FAA Administrator Bryan Bedford calls this the “third wave” of modernization. At the Modern Skies Summit on April 21, 2026, he highlighted digital twins of the National Airspace System, built from over 20 years of data. These models enable predictive analytics to deconflict traffic and cut block times, promising less fuel burn and lower costs for airlines. “If we do this correctly — we believe the early indications are that we can — we’ll actually reduce costs for our stakeholders,” Bedford said, as covered by FedScoop. Peraton, the prime integrator, deploys agentic AI to manage the program, analyzing 5.7 million records and flagging risks early. Justin Ciaccio of Peraton noted: “Risks are brought up before they become crises. Schedules are stress tested before they break.”
But here’s the rub. Duffy insists humans stay in charge. “AI is a tool, but we do not replace humans in how we manage the airspace. Am I gonna replace a controller and have AI manage the airspace? The answer to that is hell no, that’s not gonna happen,” he told CBS, echoed in Just the News. Controllers won’t vanish. AI assists on scheduling and trajectories. Yet funding lags. The AI customization lacks allocation now. Duffy joked they scrounged “couch cushions” for vendor talks. The White House’s fiscal 2027 budget offers $4 billion—far short of the $20 billion ask.
Cynics point to AI’s track record. Hallucinations plague large language models, generating false info with confidence. Failures abound in scheduling and predictions. One study showed AI botching staff rosters; another struggled with traffic forecasts, per links in Futurism. Even vending machines turned cartel under AI pricing—prices spiked in coordination. Scale that to 45,000 daily flights. A glitch could cascade. Weather shifts. Equipment fails. Human intuition catches nuances machines miss.
Peraton’s work shows promise, though. Daily scans of 1.5 million records spot supply gaps before delays hit. Still, no off-the-shelf platform fits. FAA builds custom. Air Space Intelligence praised Duffy and Bedford’s urgency on X, calling it a bias for action. Palantir’s edge? Proven government contracts, from defense to certification monitoring.
Delays cost airlines $100 billion yearly, by some estimates. SMART targets that. Predictive tweaks—mere minutes—could smooth peaks at hubs like Atlanta or Chicago. Controllers gain breathing room. But integration demands trust. FAA’s aging infrastructure, floppy disks included, begs upgrade. This $12.5 billion down payment funds fiber, digital displays, and now AI. Recent X posts buzz with it; FedScoop’s coverage drew eyes to the summit.
Challenges loom large. Staffing shortages plague controllers—retirements outpace hires. AI eases burden, not bodies. Unions watch warily. Europe modernized decades ago; America plays catch-up. Success hinges on execution. Vendors compete fiercely. Palantir’s data prowess versus Thales’ aviation heritage. ASI brings startup speed.
Bedford eyes operation later this year. Pilots and airlines wait. Boom times in travel strain the system. AI could unlock capacity without new runways. Or it falters. History favors caution. Controllers’ split-second calls save lives daily. Algorithms predict. Humans decide. For now.
The push accelerates. Duffy’s team moves fast. Congress must follow with cash. Skies stay human-led. But AI whispers changes. Watch closely.
WebProNews is an iEntry Publication