US Federal News Bureau
Written by: Tathagata Sen
Updated 2:18 PM EST, November 16, 2026

Questions are emerging over the reliability and safety oversight of the Federal Aviation Administration’s (FAA) new AI-supported airspace management system, following a November 13 nationwide airline ground stop caused by communications problems at multiple air traffic control centers, according to Associated Press report published November 16.
As per the report, the FAA has said the disruption was unrelated to the system, but the incident has renewed scrutiny of its role in decisions affecting air traffic.
The scrutiny comes days after the FAA expanded its $875 million Strategic Management of Airspace, Routes and Trajectories (SMART) platform nationwide on November 10, bringing all 22 air traffic control centers under the system.
It’s important to note that in October, almost two dozen Democrats on the House Transportation and Infrastructure Committee have asked the Government Accountability Office (GAO) to review the FAA’s use of AI in air traffic tools.
SMART consolidates 200 data streams, including weather patterns, flight paths, traffic flows and controller staffing metrics, to help specialists anticipate disruptions and recommend changes, according to the FAA’s announcement.
For chief data officers (CDOs), SMART illustrates the AI governance demands that accompany AI systems operating across large, interconnected environments. Integrating hundreds of data feeds requires reliable source information, consistent validation and clear responsibility for resolving discrepancies.
The system also highlights the importance of documenting how AI-generated recommendations are produced, communicating their limitations and preserving human authority over consequential decisions.
These requirements are particularly important in safety-critical environments, where incorrect or misleading outputs can have immediate consequences. Establishing clear AI governance roles and accountability is therefore central to deploying such systems responsibly.