When the Air Force unveiled the F-47 designation for its sixth-generation stealth fighter in April 2025, the most striking detail was not the jet itself. It was everything flying alongside it. The F-47, developed under the Next Generation Air Dominance program, is designed to operate as a command node at the center of a formation of autonomous drone wingmen. Those drones will push into hostile airspace ahead of the pilot, absorbing enemy fire, mapping air defense networks, and feeding targeting data back to the manned cockpit in real time.
The concept represents the most significant shift in American air combat architecture since the introduction of stealth. And as of mid-2026, the program is moving from PowerPoint slides toward hardware, with major questions about scale, autonomy, and cost still unresolved.
A fighter that commands a fleet
The Pentagon’s official announcement, made by President Donald Trump and Defense Secretary Pete Hegseth, described NGAD not as a single aircraft but as a “family of systems.” That family includes the manned F-47 stealth fighter, autonomous drone wingmen developed under the separate Collaborative Combat Aircraft (CCA) program, and additional networked elements the Defense Department has not fully detailed in public.
The framing matters. The F-47 is not simply a faster or stealthier replacement for the F-22 Raptor. It is built to sit at the hub of a distributed combat network, directing unmanned platforms that operate ahead of and around the manned jet. Each drone in the formation serves a distinct tactical role: some carry sensors to extend the pilot’s situational awareness, some carry weapons to increase the formation’s firepower, and some exist primarily to provoke enemy air defenses into revealing their positions by drawing radar locks and missile shots.
Analysis published by the Lieber Institute at the U.S. Military Academy at West Point lays out the operational logic in concrete terms. Autonomous wingmen are intended to fly ahead of manned fighters, probing air defenses by triggering engagements that force adversary radars and missiles to react. The goal is twofold: drain the enemy’s expensive interceptor inventory on cheap, replaceable platforms, and build a real-time picture of the threat environment without putting a pilot directly in the kill chain.
That approach, sometimes called “loyal wingman” doctrine, inverts traditional air combat economics. A modern surface-to-air missile can cost several million dollars. If it is spent destroying a drone worth a fraction of that amount, the exchange favors the side fielding the drones. Every interceptor fired at an unmanned platform is one fewer available to target the F-47 or other high-value aircraft. The math works only if the drones are affordable at scale and the pilot can coordinate them effectively from a single cockpit.
The CCA program: building the wingmen
The autonomous drones that will fly alongside the F-47 are being developed under the Air Force’s Collaborative Combat Aircraft program, which is separate from but tightly linked to NGAD. In 2024, the Air Force awarded Increment 1 CCA contracts to Anduril Industries and General Atomics Aeronautical Systems, selecting two companies to build competing prototypes of the first operational drone wingman.
Former Air Force Secretary Frank Kendall discussed a fleet-wide aspiration of roughly 1,000 CCAs during congressional testimony, a figure that has circulated widely in defense reporting. That number reflects a planning target for the total CCA inventory across the Air Force, not the count a single F-47 would command on a given sortie. In practice, a single manned fighter might direct a handful of CCAs at a time, with the broader fleet distributed across squadrons and theaters.
The distinction matters because it shapes expectations about what the F-47 pilot actually does in the cockpit. Managing four to six drones in a contested environment is a demanding but plausible task with advanced automation and interface design. Commanding hundreds simultaneously from one aircraft is not what the Air Force has described. The pilot’s role is closer to a quarterback calling plays for a small team than an air traffic controller managing a swarm.
Why a lone stealth jet is no longer enough
The NGAD architecture reflects a hard-earned lesson from two decades of watching adversary air defenses evolve. Modern integrated air defense systems, particularly those fielded by China and Russia, fuse radar, infrared, and electronic intelligence across multiple dispersed sites. A single stealth aircraft, no matter how advanced its radar cross-section reduction, faces growing difficulty slipping through that kind of layered shield undetected.
A distributed formation of manned and unmanned aircraft attacks the problem from several angles at once. Drones can fly riskier routes, operate closer to known missile batteries, and accept higher probabilities of loss to expose gaps the F-47 can exploit. If an enemy radar lights up to track a CCA, the manned fighter and other sensors in the network can geolocate that emitter and plan a strike before the radar operator can relocate.
Within this architecture, the F-47 functions less as a traditional dogfighter and more as a flying command post. Its pilot manages data flows from multiple drones, prioritizes threats, assigns targets, and decides when to commit weapons. The aircraft’s onboard computing power and secure communications links are as central to its design as its stealth shaping or engine performance. Information dominance, not just airframe superiority, is the point.
Unresolved questions that will define the program
For all the conceptual clarity, the F-47 and CCA programs carry a long list of unanswered questions that will determine whether the vision translates into battlefield capability.
Autonomy levels. The Lieber Institute’s analysis examines the legal and ethical dimensions of drones that deliberately invite radar locks and missile shots. But neither that analysis nor the Pentagon’s announcement specifies the command protocols governing when a drone can engage a target on its own versus when it must wait for a human decision. The distinction between “human-on-the-loop” oversight (where the machine acts unless a human intervenes) and “human-in-the-loop” control (where the machine waits for explicit permission) has enormous consequences for reaction speed, legal accountability, and escalation risk. Even within a single mission, drones might shift between modes depending on the threat environment, and the rules governing those transitions have not been disclosed.
Cost at scale. The loyal wingman model only works if the drones are genuinely expendable, meaning cheap enough that losing them in combat is an acceptable trade. Early CCA cost targets have been discussed in the range of roughly $30 million per unit, which is far less than a manned fighter but not trivial. The Air Force has not released detailed unit-cost estimates or procurement projections that would allow independent analysts to stress-test the economic assumptions. If CCA costs creep upward during development, as costs often do in defense acquisition, the cost-exchange math that underpins the entire concept could erode.
Timeline. The Pentagon has not disclosed a first-flight date for the F-47, an initial operational capability target, or a detailed production schedule. Budget pressures and a 2024 restructuring of the NGAD program have raised questions about pace. Without firm public milestones, outside observers cannot assess whether the program is on track or facing the kind of schedule growth that has plagued earlier fighter efforts like the F-35.
Adversary countermeasures. If the operational concept works as described, rival militaries will have strong incentives to adapt. Potential countermeasures include electronic jamming designed to sever the data link between the F-47 and its drones, counter-swarm systems that target cheap unmanned platforms in bulk, and deceptive radar tactics intended to feed misleading data back through the drone network. China’s People’s Liberation Army has invested heavily in electronic warfare and counter-drone capabilities, and any assessment of the F-47’s effectiveness must account for an adversary that is actively working to defeat the concept.
What the evidence actually supports
Two tiers of evidence underpin the public understanding of this program, and they differ in weight. The Defense Department’s own announcement is a primary source confirming the F-47 designation, the NGAD program name, the inclusion of manned and unmanned components, and the description of a connected system of platforms. Those facts are on the record from the institution building the aircraft.
The Lieber Institute’s analysis is an expert interpretation of the operational concept, not a procurement document or flight-test report. Its value lies in articulating what “draw fire” and “extend sensor reach” mean in practical terms: drones flying ahead, absorbing radar locks, and mapping threat positions so the manned fighter can maneuver with better information. That analysis is rigorous and published by a respected institution, but it reflects doctrinal reasoning rather than empirical performance data from actual drone-wingman formations.
No publicly available simulation results, flight-test reports, or after-action reviews confirm that autonomous wingmen perform as the concept promises. The Air Force has conducted classified exercises reportedly involving early loyal wingman prototypes, but the open-source evidence base consists of official announcements, congressional testimony, and expert analysis rather than verified performance data.
The strongest reading of the evidence: the Pentagon has committed to building a sixth-generation fighter designed to command autonomous drones. The operational rationale is sound and well articulated by credible analysts. But the specific numbers, timelines, and performance claims circulating in public discussion often outrun what the Defense Department has actually confirmed. Whether the F-47 and its CCA wingmen deliver on their promise will depend on engineering decisions, budget battles, and adversary responses that are still playing out behind classified doors.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
