By 2030 the character of air superiority will be defined less by a single dominant manned fighter and more by integrated formations of crewed platforms and purpose-built unmanned systems. The United States Air Force’s Collaborative Combat Aircraft experiment and the emergence of a genus approach to attritable platforms are accelerating that shift. In April 2024 the Air Force narrowed Increment 1 CCA awards to two firms, signaling a transition from conceptual experimentation to production representative test articles.

Three converging technical trends will determine whether unmanned systems can credibly contest or control contested airspace by 2030: scalable, mission-focused airframes; resilient autonomy and distributed control; and survivable, low-signature networking in a high-electronic-warfare environment.

Scalable airframes and the attritable economics

A practical path to mass is already visible. The Air Force and industry are pursuing lower-cost, repeatable chassis that can be specialized with sensing, weapons, or communications kits. General Atomics’ XQ-67 family was presented publicly as a second generation of autonomous collaborative platforms and flew its prototype in early 2024, illustrating the genus concept where a common core permits rapid replication of mission variants.

Parallel international work points to two contrasting procurement philosophies. Australia and Boeing are operationalizing the MQ-28 Ghost Bat as a loyal wingman produced in dedicated facilities, targeting an inexpensive, interoperable companion that augments fifth-generation fighters. The Boeing facility announcement in 2024 illustrated industrial commitment to producing unmanned wingmen at scale.

If the United States scales CCAs to the planning figures discussed by senior leaders, force structure could shift dramatically. Secretary-level planning assumptions made since 2023 treated a notional 1,000 CCAs as a basis for organizational analysis, while more near-term expectations by Air Force leadership anticipated roughly 100 CCAs in the FYDP window for Increment 1. Those planning math and procurement tempos matter because operational mass will be required to offset attrition, sensor saturation, and countermeasures.

Resilient autonomy and human-machine teaming

Autonomy is evolving along three axes: robustness to lost communications, explainable and verifiable decision logic for lethal functions, and social navigation to operate inside formations with human pilots. Publications and program briefs in 2024 emphasized human-machine teaming as the dominant operational concept, with autonomous platforms executing delegated tasks under pilot supervision rather than operating purely independently. This approach reduces legal and operational risk but increases dependence on integrated mission software and reliable command and control.

Research in machine learning for air combat shows potential for autonomous maneuvering, but also highlights brittleness when simulated environments diverge from contested, jammed real-world conditions. Several academic efforts in 2024 demonstrated promising dogfight policies in simulation, but those experiments also confirmed the need for substantial validation in degraded communications and contested sensing scenarios before autonomy can be entrusted with high-stakes, independent engagement.

Networked operations and the EW vulnerability

The operational value of loyal wingmen depends heavily on datalinks, airborne C2, and federation with airborne sensors such as AWACS or airborne battle managers. CSIS analysis in 2024 framed command and control options for CCAs and warned that how the Air Force architected those links will determine effectiveness. If links are fragile under jamming and spoofing, humans will be forced back into direct control roles or platforms will default to conservative behaviors that blunt their tactical advantage.

History and near-term combat experience show the risk. When a major power loses control of a complex unmanned system in a contested zone, the result can be catastrophic both operationally and politically. The October 2024 loss of a prototype heavy stealth UCAV in eastern Ukraine exposed how fragile autonomy and comms can be in the wild, and how much intelligence value adversaries can extract from a single wreck. That episode underscored the need for designs that accept partial autonomy and local autonomy modes rather than dependence on continuous, high-bandwidth links.

Adversary approaches: stealth UCAVs versus attritable wingmen

Two strategic pathways are visible among advanced military powers. One approach emphasizes a small number of high-end, stealthy UCAVs intended to penetrate deep defenses with large internal payloads. The Chinese GJ-11 Sharp Sword is an example of this path and by 2024 had been observed in expanded testing patterns that suggested an eventual role as a penetration and strike asset or as a wingman for Chinese stealth fighters.

The US model has trended toward cheaper, mass-produced CCAs that operate under tighter human supervision and rely on swarming or distributed effects instead of single-platform stealth. Both approaches are plausible routes to air superiority. The stealth-UCAV approach demands maturity in low-observable design, reliable autonomous navigation in denied environments, and expensive low-rate production. The attritable wingman model trades single-platform survivability for numbers, flexibility, and lower per-unit cost. Cost tradeoffs and doctrine will decide which mix dominates a given theater.

Operational implications by 2030

  • Distributed air dominance. Expect combat formations to include a mix of crewed fighters, high-end UCAVs where available, and large numbers of specialized CCAs for sensing, jamming, decoying, and strike. This distribution changes kill chains and attribution calculus.

  • Tempo and logistics. Mass attritable fleets will stress sustainment pipelines. Production rates, spare parts, software updates, and range infrastructure will become as strategically important as missile magazines and sortie generation rates.

  • Command doctrine. Tactical doctrine will need to codify degrees of autonomy, fail-safe behaviors, and escalation control. Human operators will remain the legal and operational decision nodes for lethal action, but delegation boundaries will expand.

  • Vulnerability to EW and supply chain attacks. Resilient datalinks, diverse sensor fusion, and hardware provenance controls will be force multipliers. The 2024 recoveries of prototype systems in contested zones showed the intelligence and supply-chain risks of exposed platforms.

What must change in acquisition and training

1) Buy software defined, modular systems. If a common chassis can accept evolving compute stacks, sensor pods, and EW kits then upgrades are faster and survivability improves. The XQ-67 messaging around a genus chassis is a good template.

2) Fund resilient autonomy testbeds. Realistic, contested-spectrum test ranges and red-teaming of autonomy under EW conditions are essential. Simulation-only proofs will not expose failure modes of machine learning models in the real world. Academic and industry work in 2024 underlined this shortfall.

3) Harden the supply chain and deny intelligence windfalls. Hardware provenance, counterfeit mitigation, and rapid-erasure measures for recovered assets must be part of design and doctrine. The October 2024 crash highlighted how much can be learned from a single recovered airframe.

4) Rethink training and doctrine for mixed teams. Pilot training must become as much about operating swarm managers and information supervisors as flying the jet. Exercises and wargames should prioritize human-autonomy interactions and edge-case contingencies. CSIS work on C2 options offers a menu of tradeoffs to evaluate.

Conclusions and realistic timelines

A credible unmanned contribution to air superiority by 2030 is plausible but conditional. Where programs achieve mass produced, affordable CCAs with robust datalinks and verified autonomy, they will alter operational math and provide commanders with new options for risk distribution. The balance will vary by region and adversary capability. Against an opponent fielding mature stealth UCAVs and integrated EW layers, the attritable wingman model will need greater emphasis on distributed autonomy and local decision making.

By 2030 the most likely outcome is a blended architecture. Crewed sixth-generation fighters, if fielded, will serve as mission commanders in contested environments and will be accompanied by multiple classes of unmanned platforms: small expendables for saturation and deception, medium-range CCAs for sensing and localized strike, and a limited number of heavyweight UCAVs for specific deep strike missions. The precise mix will be decided by budgets, attrition experience, and how swiftly autonomy proves trustworthy under contested conditions. The technical and doctrinal choices made in the next five years will therefore determine whether unmanned systems deliver strategic advantage or simply become expensive targets.

Practical takeaway: prioritize resilient autonomy, network survivability, and industrial capacity for mass production. The programs that integrate those three elements will be the ones that matter in 2030.