FAA Bets $875M on AI to Spot Flight Conflicts Two Hours Out

A quiet contract award in late June is more interesting than the usual aviation-AI demo reel, because it puts real federal money behind a predictive system in the operational stack. According to reporting from the Financial Times and corroborated by Bloomberg and The Air Current, the FAA awarded a 12-year, $875 million contract to Air Space Intelligence to build SMART, the Strategic Management of Airspace Routing Trajectories system, beating Palantir and Thales for the work.

The design pitch is straightforward. Today's controllers work with a planning window that runs about 15 minutes ahead. SMART, as described in the contract, would stretch that window to roughly two hours by ingesting weather, traffic and flight plan data faster than any human team can manage. Transportation Secretary Sean Duffy and FAA Administrator Bryan Bedford announced the award on June 22, 2026, with the Boston startup beating two much larger competitors. ASI separated itself in part by walking in with a platform already running at scale: its Flyways AI system supports flight routing for Alaska Airlines, Delta and United, and the company claims it touches more than 40 percent of US air traffic.

Why it matters: the agency is running roughly 3,500 controllers below its target staffing level and is trying to hire at least 8,900 more by 2028. Predictive AI is being positioned as augmentation, not replacement, because human controllers still make real-time judgment calls and pick up on stress in a pilot's voice. But once an agency commits 12 years and close to a billion dollars to a single predictive platform, the operating model around it starts to bend whether the agency intends it to or not.

The honest caveat is that the public reporting does not describe what SMART looks like in a tower on day one, which centers go first, how its recommendations are governed when they disagree with a controller, or who carries liability if a forecast misses. Take the two-hour figure as the contract's design target, not a deployed capability. The reporting also does not say what Palantir and Thales walk away with from the broader $32.5 billion modernization push.

If SMART delivers on even part of its design, the winners are not only ASI and the FAA. Airlines already running Flyways get tighter coupling with airspace planning, and regulators outside the US get a reference deployment to point at when they start their own predictive ATC procurements.