Jeppesen ForeFlight is the latest aviation company to introduce an artificial intelligence-powered platform as the industry seeks to incorporate more AI into various aspects of flight and operations management.
Jeppesen on Monday unveiled three AI agents within its Airflow “aviation operating intelligence” solution, the first of hundreds it plans to introduce on the platform, which was unveiled in July. The company contrasted generic AI systems—which it said sound confident but are often wrong—with what it bills as a more trustworthy “aviation intelligence” that uses a working model of the ecosystem the agent is analyzing.
The rollout advances one of Jeppesen’s first new ventures since Boeing sold it and ForeFlight to private equity firm Thoma Bravo in April 2025. In July, it partnered with uAvionix to introduce SkyPlay, a new integration that embeds ForeFlight directly within panel-mounted avionics.
A “trip trading” agent is designed to help crew members who trade trips get “faster” and “fairer” outcomes, Jeppesen said. It said the capability will help airlines limit unplanned absences and premium pay when trades are too difficult to make.
The company’s “target mastery” agent is intended to eliminate trial and error in pairing and roster planning. It can translate planners’ targeted outcomes, cost, fatigue, robustness, and other parameters into a workable flight plan.
The third agent, “augmented ops control,” would help operators manage disruptions by evaluating and ranking all recovery options, providing reasoning for each. For example, the tool could recommend a route that provides the greatest fuel savings.
“Every recovery option, weighed and recommended in seconds, with the reasoning attached for trusted decision making,” said Brad Surak, CEO of Jeppesen, in a news release. “That is what operators have told us they need most.”
The company said it plans to expand the agents to business and military aviation customers “soon.” It is seeking commercial aviation partners for a “co-development phase” before a broader rollout planned for 2027.
How It Works
Jeppesen said the agents reason using a foundation of certified aviation data, allow users to audit their decisions, and embed within existing workflows while keeping humans at the “center of decision making.”
Customers can use prebuilt ForeFlight agents, build their own agents on the platform, deploy third-party tools such as Slack, or use some combination of the three. They can set the level of automation from fully autonomous decision-making to total human control, or somewhere in between, with the human approving, monitoring, or overriding actions proposed by the agent.
“We are leveraging decades of experience delivering data and software solutions globally, which compounds virtually every second on the 60 million flights we serve annually,” Surak said in July. “But most importantly, our AI earns trust one decision at a time.”
Per Jeppesen’s website, Airflow is built on 90 years’ worth of performance models, procedures, optimizers, aeronautical data, and other certified data from more than 200 countries and ANSPs. Agents will use the company’s aviation context graph as foundation for reasoning. They will leverage commercially available data—including nav data, charts, procedures, weather, notices to airmen (NOTAMs), traffic, and flight history—and proprietary customer data and operator rules, while being tailored to specific roles such as dispatcher or scheduler.
Rather than simply spotting patterns and basing decisions on what tended to happen in past situations, the agents are designed to reason through cause and effect. They will allow users to view their inputs, sources, logic and reasoning, expected outcomes, and constraints and alternatives considered before they choose whether to accept its decisions. Agents will also have simulation ability to anticipate disruptions before they impact operations.
Jeppesen said the agents are governed by a patent-pending AI-Safety Management System (SMS) based on the industry SMS framework.
AI In the Cockpit
The October 2026 edition of FLYING Magazine highlights some concerns that experts hold regarding AI integration in aviation. Still, many companies are forging ahead developing automated and fully autonomous tools for the cockpit and elsewhere.
For example, PlaneDNA recently introduced a tool designed to change how pilots access and learn about their aircraft through AI-assisted logbook insights. Another company, Daedalean, is developing a machine learning-based computer vision system called PilotEye that is intended to augment a human pilot’s see-and-avoid efforts.
Other companies, such as Merlin Labs, are taking things a step further. Merlin’s AI-based Pilot platform is designed to autonomously assist or conduct takeoffs, landings, and other in-flight functions. Initially, it will serve as an AI copilot, but Merlin believes it could eventually enable fully autonomous flights.
Other AI-powered systems, such as Sabre’s Mosaic or Lufthansa Systems’ NetLine/Ops ++ aiOCC, are designed to help airlines better serve their customers and optimize their operations using real-time or historical data.
Even the federal government is looking to incorporate AI in aviation. The FAA last month rolled out SMART (Strategic Management of Airspace, Routes and Trajectories) as an AI-powered tool designed to assist air traffic controllers. SMART synthesizes information from hundreds of data streams and uses it to make recommendations to controllers, which they can accept or decline.