Somewhere over the California desert, an aircraft took off without a pilot in the cockpit. The U.S. Air Force’s Collaborative Combat Aircraft program has moved from blueprints to flight testing, and Anduril Industries’ YFQ-44A drone is now part of the evaluation pipeline. A Defense Department announcement confirms that a CCA prototype has completed flight testing, though it does not name the YFQ-44A specifically. Anduril’s drone falls within the broader CCA testing structure that the release describes. The test marks a turning point in the Pentagon’s most ambitious push to pair autonomous drones with human fighter pilots in combat.
The Air Force’s goal is blunt: build affordable, expendable aircraft that can fly into contested airspace so crewed jets and their pilots do not have to. A Congressional Research Service analysis (IF12740) of the CCA program details fiscal year 2026 budget requests that reflect sustained, large-scale investment in the effort, though the brief does not publish a single consolidated dollar figure for the entire CCA line. The program has become a central pillar of how the service plans to fight in the coming decades.
From concept to concrete testing
The CCA program follows a three-phase testing structure designed to progressively raise the bar. First, the aircraft manufacturer conducts vendor-led flights under controlled conditions to prove the drone can fly safely and perform basic functions. Next, the Air Force takes over at Edwards Air Force Base in California, where testers evaluate whether the aircraft meets military standards for reliability, safety, and integration with existing fighters. Finally, the program moves to Nellis Air Force Base in Nevada for operational assessments that simulate real combat scenarios, including complex air defenses and degraded communications.
The YFQ-44A is built around a concept the Air Force calls human-machine teaming. Rather than replacing pilots, the drone is designed to extend what a single crewed fighter can accomplish. A pilot flying an F-35, for example, could direct one or more unmanned wingmen carrying sensors, weapons, or electronic warfare payloads into high-threat areas. If a drone is shot down, the financial and human cost is a fraction of losing a crewed jet and its pilot.
That cost calculus drives the entire program. The Air Force needs a larger fleet to compete with near-peer adversaries, but it cannot afford to fill that gap with more fifth-generation fighters at current price points. Autonomous drones built at a fraction of that cost offer a way to grow the force without exhausting the budget. The CRS analysis frames this as the program’s core rationale: mass and affordability in an era when the threat demands both.
The logistics challenge behind the flights
Flying a prototype is one thing. Fielding a fleet of autonomous wingmen across forward bases in the Pacific or Europe is another problem entirely. The CRS analysis notes that the Air Force is already working through basing and sustainment concepts for CCA aircraft, addressing questions that will determine whether the drones can actually be deployed at scale.
Those questions are practical and unglamorous but critical. Where will the drones be stationed? How large is the maintenance footprint? What does the spare parts pipeline look like? How will ground crews be trained to support aircraft that think for themselves? An autonomous drone that performs brilliantly in a Nevada test range but cannot be rapidly deployed from a dispersed airfield in the western Pacific loses much of its tactical value. The Air Force appears to recognize this, folding logistics planning into the program alongside flight testing rather than treating it as an afterthought.
What the public record does not yet show
For all the progress the program has made, significant gaps remain in what the public actually knows about the YFQ-44A. No official Air Force evaluation reports from Edwards AFB on the drone’s integration with manned aircraft have been released. Specific performance data from initial flights, including speed, range, payload capacity, and how well the aircraft’s autonomous systems make decisions in real time, has not appeared in any government publication.
Anduril itself has kept technical details close. The company’s public statements have emphasized the YFQ-44A’s general purpose and design philosophy rather than quantified results. Without independent Air Force test data, it is difficult to measure how the drone stacks up against competing CCA prototypes. The vendor-led testing phase, by its nature, means the manufacturer controls the conditions and the narrative around early results.
The broader question of electronic warfare resilience also lacks public answers. A near-peer fight would involve heavy jamming, GPS denial, and cyber interference. Whether the YFQ-44A or any CCA prototype can maintain reliable autonomous operations under those conditions has not been addressed in available government assessments. The later phases of testing at Edwards and Nellis are designed to stress exactly these variables, but results have not been published.
Unresolved policy questions on autonomous lethal authority
The CCA program does not exist in a vacuum. The acquisition timeline carries real uncertainty. While fiscal year 2026 budget requests confirm near-term funding, the path from prototype testing to full-rate production involves multiple decision gates where programs can be delayed, restructured, or scaled back. Congressional oversight, shifting threat assessments, and budget pressures all introduce variables that could alter the timeline.
There is also an unresolved policy dimension that could shape how aggressively the Air Force employs these aircraft even if the technology works. The public record does not clarify how much lethal decision-making authority will be delegated to autonomous systems like the YFQ-44A. Current Department of Defense policy, outlined in DoD Directive 3000.09, requires that autonomous weapons systems be designed to allow human operators to exercise appropriate levels of human judgment over the use of force. How that directive applies to fast-moving air combat scenarios, where milliseconds matter, remains a subject of active debate.
What flight testing proves and what it does not
The available evidence as of spring 2026 supports a clear but conditional picture. The Air Force has committed real money and real flight hours to the CCA concept. The YFQ-44A is flying within a testing pipeline that is structured and funded, and logistics planning is underway. Those are not trivial milestones for a program that was still largely theoretical just a few years ago.
But the most consequential judgments about the YFQ-44A’s capability, survivability, and value for money depend on data that has not yet entered the public domain. The jump from “it can fly” to “it can fight” is enormous, and the Air Force’s phased evaluation structure exists precisely because each step reveals problems that controlled vendor tests cannot. Until Edwards and Nellis assessments produce results, and until the Air Force releases at least some performance benchmarks, outside observers are left to weigh verified intent against unproven execution.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
