FAA Seeks AI for Predictive Air Traffic Management

▼ Summary
– The FAA is developing an AI system called SMART to predict flight conflicts up to two hours in advance, a major increase from the current 15-minute window.
– Three companies—Palantir, Thales, and Air Space Intelligence—are competing for the contract, each bringing distinct government, legacy, or specialized aviation AI expertise.
– The project’s urgency follows a LaGuardia Airport crash that exposed controller overwork and outdated safety systems within a broader $32.5 billion modernization program.
– The FAA’s modernization includes replacing 612 outdated radar systems and hiring 1,200 new controllers, with SMART as a central pillar of this overhaul.
– The agency’s history with slow, costly tech upgrades and workforce resistance poses a challenge to SMART’s promised rapid deployment timeline.
The Federal Aviation Administration is advancing a pivotal artificial intelligence system designed to transform national airspace management. This initiative, known as Strategic Management of Airspace Routing Trajectories (SMART), aims to extend the planning horizon for air traffic controllers from a reactive 15-minute window to a predictive two-hour timeframe. Three firms, Palantir, Thales, and Air Space Intelligence, are competing for the development contract, with Transportation Secretary Sean Duffy confirming the competitors on April 17. Further project details are expected at a press event scheduled for April 21.
SMART utilizes high-fidelity 4D modeling to forecast potential bottlenecks and scheduling conflicts before aircraft even depart. This shift from reactive to predictive air traffic management addresses a core vulnerability: the existing infrastructure, built for a lower volume of flights, depends heavily on human controllers making real-time decisions with limited forward visibility. The FAA has indicated a version of the system could be operational before the end of the year.
Each bidder brings a unique strategic advantage. Palantir Technologies leverages its extensive government contracting experience and robust data integration platforms. Its revenue, guided to hit roughly $7.2 billion in 2026, is fueled by major defense contracts and partnerships with firms like GE Aerospace. The company’s proposal centers on its proven ability to process vast operational datasets into actionable decision-support tools.
In contrast, Thales offers decades of embedded domain expertise. The European defense giant has supplied air traffic management systems to the FAA for over 85 years, with its equipment present in more than 99% of U. S. instrument landing systems. Its incumbent status with the existing TopSky platform provides a significant integration advantage for the SMART project.
The third contender, startup Air Space Intelligence, presents a highly focused solution. Its Flyways AI platform already manages over 40% of all U. S. air traffic through airline partnerships, employing the same 4D modeling and optimization SMART requires. ASI’s recent collaboration with Joby Aviation on integrating electric air taxis further positions it at the forefront of next-generation airspace challenges.
The push for SMART is driven by acute and recent safety concerns. The March 22 collision of an Air Canada Express flight with a fire truck at LaGuardia Airport exposed critical systemic flaws. Investigators found an overworked controller handling multiple roles and an outdated automated safety system that failed to issue an alert. This incident highlighted the growing risks of controller overwork, aging technology, and shrinking error margins as flight volumes increase.
This project is a cornerstone of a broader $32.5 billion modernization program. Congress has allocated $12.5 billion, with an estimated $20 billion more required to complete the overhaul. Concurrent efforts include replacing 612 obsolete radar systems, moving the NOTAM system to the cloud, and an aggressive controller hiring campaign, with nearly 1,200 recruited in fiscal 2026 alone. FAA Administrator Bryan Bedford has prioritized SMART within this agenda.
External pressure also comes from DOGE, the Department of Government Efficiency, which has conducted evaluations of air traffic control facilities to recommend rapid safety upgrades. While DOGE is scheduled to cease operations on July 4, its successor is expected to continue engagement. A parallel effort, Project Lift, is channeling FAA funds toward upgrading network communications infrastructure.
The contract competition underscores three divergent models for government AI procurement. Palantir offers a versatile, security-cleared platform adaptable across agencies. Thales provides unmatched domain-specific expertise and an installed hardware base. Air Space Intelligence delivers a purpose-built, commercially proven aviation AI solution. The FAA’s track record, however, invites skepticism; the protracted and costly NextGen modernization program serves as a cautionary tale regarding timelines and workforce resistance to new automation.
The implications extend far beyond a single contract award. For Palantir, a win would cement its expansion into a major civilian agency and support its lofty market valuation. For Thales, it would modernize a foundational client relationship. For Air Space Intelligence, it would validate its commercial approach on the national stage.
Ultimately, the stakes concern the integrity of the U. S. air traffic control system itself, which coordinates about 45,000 daily flights across the world’s most complex airspace. The system is strained by aging technology, staffing shortfalls, and increasing traffic that tests historically strong safety margins. SMART represents a critical bet that AI-driven predictive analytics can bridge the gap between the system’s original design and current demands. The central question remains whether any of the three competing firms can deliver a functional solution at the accelerated pace the FAA’s safety imperative now demands.
(Source: The Next Web)

