The Department of the Air Force’s Collaborative Combat Aircraft program is no longer a distant thought experiment. In April 2024 the Air Force exercised option awards that narrow the first production-representative phase to two teams: Anduril and General Atomics Aeronautical Systems (GA-ASI). That award moves CCA from concept and experimentation toward hardware that must be integrated, tested, and — crucially — sustained at scale.

What the Air Force picked and why it matters

The April decision follows a January 2024 round in which five companies were given initial development contracts to mature designs: Anduril, Boeing, GA-ASI, Lockheed Martin, and Northrop Grumman. The service’s intent at that stage was explicit: capitalize on industry experimentation, then down-select to the most promising designs for production-representative test articles. That process recognizes the program’s fundamental aim — affordable mass paired with meaningful autonomy.

That aim is not abstract. Secretary Frank Kendall and other senior leaders have framed CCA as a way to deliver “intelligent mass”: lower-cost, attritable or semi-attritable unmanned aircraft that multiply the combat effect of crewed fighters like F-35 and the future NGAD family. Kendall has offered a planning figure of roughly 1,000 CCAs based on two CCAs per NGAD and two per a slice of the F-35 fleet, although he stressed that was a planning assumption and not a program of record. He has also told Congress he views a per-aircraft cost on the order of $25 million to $30 million as an upper bound for affordability. Those numbers are the arithmetic that drives everything from industrial planning to basing and logistics assumptions.

Budget and schedule realities

The Air Force folded CCA into its FY 2025 modernization math and publicly tied meaningful funding to the program’s early increments. Pentagon and Air Force budget materials in early 2024 highlighted a multi-hundred-million-dollar commitment to CCA development in the near term and projected substantial FYDP spending to build the program out. The service has signaled an aggressive timeline: design maturation and test-article construction in the mid-2020s leading to a production decision around FY2026 and an initial operational capability later in the decade, contingent on testing and integration outcomes. Those schedule targets are ambitious and depend on getting autonomy, propulsion, mission-systems integration, and logistics right on first tries.

What the prototypes already tell us about design choices

One clear trend is the “genus and species” approach to platforms. The General Atomics XQ-67A demonstrator, which made a public first flight in February 2024, is explicitly framed by AFRL as a second-generation collaborative platform that proves a common chassis concept. That approach is attractive because it reduces per-unit engineering cost by sharing major subsystems across variants and lets mission-specific kits define role profiles such as sensing, electronic warfare, or weapons carriage. But it also concentrates risk: a common chassis means a common vulnerability vector for sustainment, cybersecurity, and airworthiness.

Autonomy, human control, and operational doctrine

CCA ambitions rest on autonomy that does more than fly straight lines. The service expects human supervisors to task groups of CCAs and to retain authority over use of force, but senior leaders have publicly acknowledged the speed-of-decision problem in high-end air combat and the need for autonomy that can act at machine timescales. That tension matters because it drives system architecture choices: autonomy stacks must support certified, auditable behaviors for contested sensing and weapons employment while also being robust to communications loss, spoofing, and cyber attack. The Air Force and DoD statements in 2023–24 indicate the intention to pair autonomy with procedural human oversight and constrained rules of engagement, but the limits of that model are operational, legal, and ethical as much as technical.

Technical and acquisition pinch points

From a systems engineering perspective the program has five chokepoints that will determine success or failure:

  • Autonomy assurance and verification: proving complex tactical behaviors across edge cases at scale is hard. Simulation helps, but distributed real-world sensing, contested electro‑magnetic environments, and degraded comms require validation regimes far beyond single-vehicle flight tests.

  • Communications and networking: CCA concepts assume resilient, low-latency datalinks between crewed aircraft, CCAs, and off‑board nodes. Those links must be hardened, spectrum‑efficient, and survivable under jamming and cyber attack. Failure modes here are mission critical.

  • Software modularity and integration: the service needs an open, standards-based mission system and modular autonomy “stack” so different vendors and mission kits can interoperate without re-certifying whole aircraft each increment. The alternative is vendor lock and spiraling integration costs.

  • Production economics and supply chain: to reach planning quantities the program must escape low-rate aerospace economics. That demands design-for-manufacture, dual sourcing of critical components, and mechanisms to avoid single points of failure in the supply chain.

  • Test, evaluation, and safety oversight: certification processes for weaponized autonomy will have to evolve. Current acquisition pathways and test ranges are not optimized for rapid, iterative autonomy testing at scale. The Air Force’s decision to fund production‑representative test articles is a recognition of that gap.

Industry and force-structure consequences

Two firms moving into the next phase does not mean only two providers in the final inventory. The structure the Air Force is using preserves a broad industrial partner pool while incentivizing rapid progress by rewarding designs that reach production readiness quickly. If CCA meets cost and performance targets, it will reshape force structure by adding numerically large, lower-cost nodes to combat formations and by changing pilot‑centric tactics to human‑supervised teaming. That will place new demands on training pipelines, maintenance units, and expeditionary basing concepts. It will also shift how capability is measured: attritability and lifecycle cost may trump per-unit capability in some mission sets.

Risks the program must manage now

  • Over‑specification. The temptation to make the CCA a mini‑fighter will blow the cost curve. The point of CCA is careful capability for a price. Requirements should be decomposed by mission kit, not baked into every airframe.

  • Cyber and supply chain vulnerability. A common chassis and common autonomy stack reduce cost but increase systemic risk unless mitigations such as diverse suppliers, secure boot chains, and independent verification are enforced.

  • Doctrine and legal calculus. The Air Force must define operational rules for loosing autonomous effects in contested environments and build the audit trails and accountability mechanisms to support them. Congressional and international scrutiny will follow as capabilities mature.

What to watch next

Short term: look for production‑representative test articles to enter ground and flight test campaigns and for the Air Force to publish concept of operations material for CCA basing and employment. Mid term: expect tougher questions from Congress on affordability, sustainment, and certification as prototype test data arrives. Long term: if the technical and acquisition risks are managed, CCAs will be the lever that changes the arithmetic of air combat from a few exquisite platforms to a fight that trades quality and quantity in a new balance.

Practical recommendations for program managers

1) Prioritize open mission-system interfaces and third-party interoperability testing now. Without it integration will become the long pole.

2) Fund and expand realistic, contested-environment test ranges and digital twin capabilities to accelerate verification of autonomy under adversarial conditions.

3) Insist on dual sourcing for critical components and separate mission‑kit supply chains from airframe supply chains to reduce systemic risk.

4) Coordinate early with legal, ethics, and operational communities so doctrine and certifiable behaviors evolve with the platforms rather than lag behind them.

Conclusion

The April 2024 down‑select to Anduril and General Atomics is a watershed, not because it picks winners by itself, but because it moves the CCA program into a phase where engineering, test, and logistics must prove the concept in the harsh realities of integration and sustainment. If the program can hold to its affordability targets while delivering autonomy that is verifiable, survivable, and interoperable, CCAs will be a disruptive multiplier for U.S. airpower. If not, they will be expensive prototypes that underscore how hard it is to scale autonomy in combat aviation. The technical and institutional choices the Air Force makes over the next 24 months will determine which of those futures becomes reality.