The Federal Aviation Administration is about to hand a big piece of its air traffic problem to software. According to TechCrunch AI, citing a Wall Street Journal report, the FAA will soon launch SMART, an $875 million AI platform built to help air traffic controllers manage workloads and route flights more safely. The contract runs 12 years, and the first rollout lands in the Washington, D.C. metro area before spreading to other regions.
The vendor is Air Space Intelligence, a firm that’s been quietly selling flight-planning AI to airlines for years. Now it’s got the government as a customer.
What SMART actually does
SMART stands for Strategic Management of Airspace, Routes, and Trajectories. Per a one-page FAA readout quoted by TechCrunch AI, it’s “a cloud-based platform system that enhances existing FAA air traffic management systems.” Note the word “enhances.” This isn’t replacing the radar screens or the controllers behind them. It sits on top.
The system pulls in “airline schedules, weather, airport capacity, airspace conditions, and operational constraints to predict traffic flows and identify potential conflicts before they occur.” In plain terms:
- It ingests a pile of live and scheduled data.
- It forecasts where planes will be and where congestion will build.
- It flags conflicts early so humans can reroute before things get tight.
That’s decision support, not autonomy. Controllers still make the calls. The software just tries to give them a longer runway to make them.
Why the FAA is doing this now
The short answer is staffing. The FAA has been short on controllers for years, and TechCrunch AI notes the causes are varied: training bottlenecks, retirements, burnout from mandatory overtime, and a pipeline that never caught up after the pandemic pause. Earlier this year the agency announced a “bold, new” hiring plan it says will “erase the longstanding staffing shortage.”
Hiring takes years, though. A new controller needs two to three years of training before they’re certified at a busy facility. Software can ship faster than that. So the strategy has two tracks: more people eventually, better tools now.
There’s a third track too. The government is separately running a broad modernization push for the nation’s aging air traffic infrastructure, some of which still runs on decades-old hardware. SMART being cloud-based is a small tell about where that effort is heading.
What stands out here
Three things worth paying attention to:
- The money is real. $875 million over 12 years is roughly $73 million a year. That’s not a pilot. That’s a commitment to one vendor for a system that will become load-bearing.
- Prediction beats reaction. Most legacy ATC tooling shows you what’s happening now. SMART’s pitch is forecasting conflicts before they happen. If the predictions are good, that’s the difference between a smooth reroute and a ground stop. If they’re bad, controllers learn to ignore the alerts, which is worse than no alerts at all.
- D.C. first is a stress test. The Washington airspace is one of the most restricted and congested in the country, with three major airports and heavy military and government traffic. Starting there says the FAA wants proof under pressure, not a cushy demo.
The bigger picture for AI
This is what “AI in critical infrastructure” looks like in practice. Not a chatbot, not a model that replaces workers. A predictive layer stitched into an existing system, with humans firmly in the loop and a government readout that goes out of its way to say “enhances.”
It’s also a signal to the industry. Air Space Intelligence isn’t a household name. A 12-year federal contract makes it one in aviation circles overnight, and it tells every other vertical AI startup that agencies are ready to buy if the product bolts onto what already exists rather than asking for a rip-and-replace.
What to watch
- Rollout timing. The report says “soon” for D.C. Expansion to other regions will depend on how the first deployment performs.
- Controller adoption. Tools like this live or die on whether the people using them trust the output. Watch for union feedback from NATCA.
- Accuracy claims. Expect the FAA to publish metrics on conflicts predicted versus conflicts avoided. If those numbers don’t show up, that’s a story in itself.
The FAA is betting that software can buy time while it rebuilds its workforce. The next 18 months in D.C. airspace will tell us if that bet pays. Full details are in the TechCrunch AI report.