Free Intensive Nov 21–22Free 2-day hands-on Vibecoding Intensive on Nov 21–22 at 16:00 MSK (Sat–Sun) · Live Cohort 3 starts Nov 23Join Free →Cohort 3 (Nov 23)
ReleaseInsight · On RadarArs Technica→ original

FAA rolls out SMART AI advisor nationwide: 10 minutes and 80 gallons saved per flight

The U.S. Federal Aviation Administration (FAA) has rolled out its SMART AI advisor at all 22 air traffic control centers. It predicts congestion in the sky and suggests routes and landing order to controllers. According to the FAA itself, 180 flights over Washington saved an average of 10 minutes and nearly 80 gallons (about 300 liters) of fuel each. There is no independent verification of these figures, so the real test is still ahead.

AI-processed from Ars Technica; edited by Hamidun News
FAA rolls out SMART AI advisor nationwide: 10 minutes and 80 gallons saved per flight
Source: Ars Technica. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceArs Technica ↗
Editor-in-Chief
Key metric #1The system costs $875 million and runs in the cloud.
Key metric #2The FAA's team reviews the recommendations, and a controller has the right to reject them.
Key metric #3The air traffic controllers' union (more than 15,000 members) did not take part in development.

When airports fill up with delays, the usual culprits are thunderstorms, a shortage of controllers, and a poor landing sequence. All of these can be modeled in advance, and that is the bet the U.S.

Federal Aviation Administration, or FAA, is making. The agency has expanded its SMART AI system from one region to the entire National Airspace System, and it now runs at all 22 air route traffic control centers in the country. A center is a regional control facility that directs aircraft flying between airports.

According to the FAA, in the Washington area the system helped reroute 180 flights, and each one saved an average of 10 minutes of flight time and nearly 80 gallons of fuel, or about 300 liters.

What SMART does and who makes the decisions

According to the FAA's description, SMART is a cloud-based system. It pulls together weather, aircraft routes, traffic density, and data on how many controllers are actually on shift. From that, it predicts where a bottleneck will form and issues three kinds of suggestions: how to schedule flights, which route is optimal, and the order in which to accept arriving aircraft.

Think of a navigation app that sees not only the traffic jam now, but also that in an hour the intersection will be short of traffic officers. SMART builds those kinds of forecasts for the sky.

The system only advises. Its recommendations are first reviewed by a team of FAA specialists in Washington and only then passed to local facilities. Local controllers decide for themselves whether to accept the advice. In its September announcement, the agency stressed that SMART cannot be used in place of controllers or for direct control of an aircraft.

The numbers and the blind spots

SMART went live over Washington on September 21. The coverage was then extended to centers in Atlanta, Jacksonville, and Miami, and on November 10 the FAA announced a nationwide launch. The project cost $875 million. Based on the agency's statements, it expects to reproduce the pilot's results on a national scale.

These data have limitations:

  • The figures come from the FAA's own statements, and there is no independent audit yet.
  • The sample is small: 180 flights over just under two months in one region. There is also no comparison with a control group of flights that received no advisories.
  • The only public example reported by the Dallas Morning News is a cargo flight to Dallas Fort Worth Airport. The agency did not name the flight and did not say how SMART affected passenger traffic.

Who is pleased and who was left out

Heads of major airlines backed the launch. The September announcement included quotes from Delta CEO Ed Bastian and American Airlines CEO Robert Isom. The benefit for carriers is clear: fewer delays and lower fuel consumption.

The NATCA controllers' union, which has more than 15,000 members, told Flying magazine that it was not involved in designing, testing, or deploying SMART. As a result, it cannot assess the system's impact on staff. For a tool whose suggestions controllers will see every shift, this is a serious gap: the people who accept or reject the advice did not take part in building it.

What follows from this

For those building AI systems in critical industries, SMART offers a working model of deployment. The AI makes a recommendation, an expert checks it, and the final operator keeps the right to refuse. This lowers the risk of error, and so far there is no basis to conclude that the system operates outside this chain.

But the case also shows a common mistake: scaling up after a short pilot and without involving the people who work with the tool. If the suggestions are not trusted, they will simply be ignored, and then neither the 10 minutes nor the 300 liters of fuel per flight will materialize. The real test will be nationwide statistics: whether the Washington pilot's results repeat in Atlanta, Miami, and Dallas, and whether anyone other than the FAA itself confirms them.

Story

Key Questions Answered

What SMART does and who makes the decisions?

According to the FAA's description, SMART is a cloud-based system. It pulls together weather, aircraft routes, traffic density, and data on how many controllers are actually on shift. From that, it predicts where a bottleneck will form and issues three kinds of suggestions: how to schedule flights, which route is optimal, and the order in which to accept arriving aircraft. Think of a navigation app that sees not only the traffic jam now, but also that in an hour the intersection will be short of traffic officers.

What are the key details on The numbers and the blind spots?

SMART went live over Washington on September 21. The coverage was then extended to centers in Atlanta, Jacksonville, and Miami, and on November 10 the FAA announced a nationwide launch. The project cost $875 million. Based on the agency's statements, it expects to reproduce the pilot's results on a national scale. These data have limitations: The figures come from the FAA's own statements, and there is no independent audit yet. The sample is small: 180 flights over just under two months in one region.

Who is pleased and who was left out?

Heads of major airlines backed the launch. The September announcement included quotes from Delta CEO Ed Bastian and American Airlines CEO Robert Isom. The benefit for carriers is clear: fewer delays and lower fuel consumption. The NATCA controllers' union, which has more than 15,000 members, told Flying magazine that it was not involved in designing, testing, or deploying SMART. As a result, it cannot assess the system's impact on staff.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Private AI Builders Community · Vibe Club

Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews

Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.

Join Vibe Club →Explore Hamidun Academy
What do you think?
Loading comments…