Auto Dynamics / Mobility Strategy

FAA Quietly Advances AI Air Traffic Control Prediction System SMART: Three Bidders and “Two Hours Ahead” Claim Cross-Checked

Based on three provided sources, the core fact that the FAA is advancing an AI-driven predictive air traffic control system called SMART (Strategic Management of Airspace Routing Trajectories) has been consistently confirmed by the first two sources; the identification of Palantir, Thales, and Airspace Intelligence as participating bidders/partners is also confirmed. The claim that the system could extend the conflict-prediction window from about 15 minutes to 1.5 to 2 hours is mentioned explicitly only in source 2. Source 3 is only weakly related to SMART and does not provide independent corroboration.

TSO brief

  • Based on three provided sources, the core fact that the FAA is advancing an AI-driven predictive air traffic control system called SMART (Strategic Management of Airspace Routing Trajectories) has been consistently confirmed by the first two sources; the identification of Palantir, Thales, and Airspace Intelligence as participating bidders/partners is also confirmed. The claim that the system could extend the conflict-prediction window from about 15 minutes to 1.5 to 2 hours is mentioned explicitly only in source 2. Source 3 is only weakly related to SMART and does not provide independent corroboration.
  • Auto Dynamics · Mobility Strategy
  • Apr 24, 2026
TSO noteEach article is checked against independent reporting. The original source links are listed with the analysis so readers can inspect the evidence directly.

Source transparency

Original reporting sources

  1. FAA quietly developing AI-enabled predictive air traffic management system - The Air Currenttheaircurrent.com
  2. Palantir, Thales, and a startup are competing to build the FAA’s predictive air traffic AI - The Next Webthenextweb.com
  3. Digital Towers Coalition To Back Remote Tech at U.S. Airports - Aviation International Newswww.ainonline.com

Top-line cross-source assessment and TSO verification conclusion:

  • Source 1 confirms that the FAA is “quietly developing” an AI-powered air traffic control software tool called SMART, and that Palantir, Thales, and Airspace Intelligence have been brought into the bidding process.

  • Source 2 confirms that the FAA is building an AI system called SMART, aimed at allowing air traffic controllers to predict and address flight conflicts up to two hours before they occur; U.S. Transportation Secretary Sean Duffy confirmed the project and the three bidders on April 17.

  • Source 3 has only a weak connection to FAA modernization and ATC-related topics and does not provide independent facts that can be used to substantiate the SMART issue.

TSO verification conclusion: The project name SMART, the FAA’s advancement of the system, and the three bidding/partner companies are all key facts that can be cross-verified by sources 1 and 2. The specific wording that the prediction window would expand “from about 15 minutes to 1.5 to 2 hours” appears only in source 2 and cannot be confirmed as a shared figure across all sources provided. Source 3 does not offer meaningful additional support.

Confirmed facts:

  1. The FAA is advancing an AI-related air traffic control system called SMART (Strategic Management of Airspace Routing Trajectories).

  2. Palantir, Thales, and Airspace Intelligence have been identified as participants in the project’s bidding or partnership process.

  3. The project is directly tied to forecasting and mitigating airspace or flight conflicts.

Main differences or points of divergence:

  1. Time-window description:

    • Source 2 explicitly states that conflicts could be predicted up to two hours in advance.

    • Source 1 only describes it as a “predictive air traffic management system” and gives no specific time window.

    • Therefore, the full claim that the window has expanded “from roughly 15 minutes to around 1.5 to 2 hours” cannot be uniformly confirmed from the provided sources.

  2. Level of public disclosure:

    • Source 1 emphasizes that the project is being developed “quietly.”

    • Source 2 says Sean Duffy publicly confirmed the project and the three bidders on April 17, and indicated more details would follow.

  3. Source 3:

    • It only has a peripheral relationship to FAA modernization and remote/ATC technology and does not provide a direct SMART report.

Background and analysis:
The FAA’s push for an AI-powered predictive air traffic tool suggests a shift in air traffic management from reacting after the fact to anticipating issues before they escalate. Based on the provided sources, the most clearly confirmed development is not a specific technical implementation, but rather that the project has reached a bidding/partnering stage and has been publicly acknowledged by the Transportation Secretary.
However, details about system performance, deployment scope, testing status, algorithmic capability, and integration with existing air traffic control workflows have not been sufficiently disclosed in the provided sources and cannot be confirmed from them. In reporting this story, it would be inappropriate to present the two-hour prediction claim as already deployed capability or to describe the three companies as final winners.

Three-source summary:

  • Source 1 (The Air Current): The FAA is quietly developing SMART, an AI air traffic control software system, with Palantir, Thales, and ASI brought into the bidding process.

  • Source 2 (The Next Web): The FAA is building SMART, with a goal of predicting and resolving flight conflicts up to two hours ahead; Sean Duffy has publicly confirmed the project and the three bidders.

  • Source 3 (Aviation International News): The article is related to FAA modernization and remote technology, but does not provide a clear independent report on SMART and has limited value here.

Conclusion:
Taken together, the three sources confirm that the FAA is indeed advancing SMART, an AI-driven predictive air traffic control project, and that Palantir, Thales, and Airspace Intelligence are in the relevant bidding or partnership picture. However, specific performance boundaries, deployment timing, and the final winner cannot be confirmed from the provided sources.

Auto Dynamics