YFQ-42A and YFQ-44A: technology, autonomy and America’s new AI combat aircraft

The decisive novelty in the YFQ-42A Dark Merlin and YFQ-44A Fury is not a single missile or an advertised range figure. It is their intended role. The US Air Force classifies them as uncrewed fighter aircraft designed to create combat effect alongside piloted jets. In June 2026 both designs received development and production contracts for the first Collaborative Combat Aircraft increment. The competition has therefore become a dual production programme, but neither aircraft should yet be mistaken for a fully fielded weapon system.

From Y prototypes to FQ production aircraft

The designation tells part of the story: Y marked the test phase, F means fighter and Q denotes an uncrewed aircraft. After the 2026 award the Air Force began referring to FQ-42 and FQ-44. General Atomics supplies Dark Merlin and Anduril supplies Fury. The label is deliberately more ambitious than the traditional remotely piloted strike drone. These aircraft are intended to support air superiority, share tactical tasks with crewed platforms and carry mission systems that can evolve without rebuilding the entire airframe.

What has actually been demonstrated

YFQ-42A entered flight test in 2025 and YFQ-44A followed. In July 2026 the Air Force reported that Fury released an AIM-120 against a digital target after a staged sequence of inert carriage and data-link evaluations. At Creech AFB, the experimental unit then generated several sorties, serviced and relaunched aircraft and integrated them with crewed platforms. These are important maturity indicators, but they do not equal completed operational test or full combat readiness.

Autonomy as replaceable mission software

The acquisition model deliberately separates air vehicle hardware from mission autonomy. A government-owned reference architecture is intended to let software from several vendors run across different aircraft. Six companies received a place in the autonomy contract pool, while selection of a primary provider is planned for 2027. The logic is compelling: avoid vendor lock, update tactics faster and compare algorithms under common standards. The hard part will be certification, cyber security, predictable behaviour and the quality of training data in a rapidly changing combat environment.

Human authority over weapon release

The AIM-120 event was sometimes described as autonomous air combat. The official description is narrower. The aircraft executed parts of the employment sequence within operator-defined parameters, but a human authorised release. The Air Force states that CCA will not autonomously employ weapons. Autonomy can cover navigation, formation keeping, sensor processing, threat response and execution of assigned tasks; it does not amount to an unrestricted machine deciding whom to attack.

Operational logic: distribute reach, mass and risk

CCA are meant to complement F-35s, the future F-47 and other command platforms. Potential roles include forward sensing, carrying additional air-to-air weapons, electronic support, deception and scouting dangerous airspace. Their value lies in distribution: an opponent must detect and engage more nodes while the crewed aircraft does not expose itself for every task. Affordable mass remains a relative term, because secure networks, autonomy software, weapons, maintenance and dispersed logistics all contribute to real ownership cost.

Why the Air Force kept both designs

The June 2026 decision did not eliminate either company. General Atomics and Anduril both moved into engineering, manufacturing and production. The Air Force says it wants more than 150 combat-capable CCA by the end of the decade and about 1,000 over the longer term. Public documents do not yet define the final split, whether the types will specialise, or how later increments will compete. Two fleets improve resilience and learning, but also create training and support burdens.

The classified core

Reliable public figures for speed, combat radius, radar signature, sensors, payload, internal carriage and unit cost are not available. Photographs show different aerodynamic solutions but cannot support credible performance calculations. Corporate artwork is not a specification sheet. The supported conclusions are broader: long-range relevance, reduced signature, modular mission systems and a smaller deployed footprint than a conventional crewed fighter are central programme goals.

Editorial assessment

YFQ-42A and YFQ-44A are no longer slideware. Flight testing, production contracts, a controlled weapon event and exercises with operational personnel make CCA one of the most tangible efforts in crewed-uncrewed air warfare. The defining test still lies ahead: delivering open software, meaningful human control and large numbers simultaneously without allowing cost and complexity to rise back toward traditional fighter levels.

Frequently asked questions

Are the YFQ-42A and YFQ-44A operational?

No. They are in development, early production and operational evaluation; full combat readiness has not been declared.

Can they decide to fire weapons on their own?

According to the Air Force, no. A human operator retains authority for weapon release.

Why produce two aircraft?

Parallel suppliers reduce dependency, preserve competition and provide comparative evidence on performance, operations and cost.

Primary sources and method

This assessment uses official government, armed-forces and manufacturer documents. Missing or classified parameters are identified explicitly.

• U.S. Air Force – Designations YFQ-42A and YFQ-44A
• U.S. Air Force – Increment 1 production and autonomy contracts
• U.S. Air Force – YFQ-44A AIM-120 test
• U.S. Air Force – Operational exercise at Creech AFB
• General Atomics – YFQ-42A programme background

Bild mit KI generiert