FAA Plans $875 Million AI System to Manage Air Traffic Congestion

Summary

The US Federal Aviation Administration plans to begin testing SMART, an AI-assisted system that advises controllers on managing congested airspace around the Washington, DC, area. Government and industry officials told The Wall Street Journal that the system could start serving the region’s three major airports as soon as September 21, 2026, before a planned expansion across the 29 million square miles of US national airspace overseen by the FAA. SMART is designed to use AI models to forecast traffic flows and identify possible conflicts from airline schedules, weather, airport capacity, and airspace conditions. Its forecasts and recommendations would be shared with the FAA, airlines, and aircraft operators through existing systems, helping them coordinate routes and departure times; the FAA says the goals include lower fuel use, better on-time performance, and faster recovery from weather and congestion. Airline concerns reportedly eased after the agency said the tool would not alter controller or airline procedures. The system is part of an $875 million, 12-year contract awarded to Boston-based Air Space Intelligence in June, which also includes a replacement for the FAA’s existing flow-management data system. The company says its separate Flyways AI platform already supports more than 40 percent of US air traffic, but the FAA has not disclosed which AI models SMART uses. The initial scope and altitude limits remain unclear. Aviation safety adviser Philip Mann argues that a limited deployment can reduce uncertainty, and says future expansion should depend on performance under real workloads and degraded data, as well as clear responsibility when predictions are wrong. The rollout comes amid aging air-traffic-control equipment, a long-running controller shortage, and heightened scrutiny after a January 2025 collision near Reagan Washington National Airport in which investigators cited tower understaffing as one causal factor.