AFRICOM’s CURTAIN CALL Swarm Cleared a Harder Test. Confirmed Kills Weren’t the Point.

U.S. Africa Command ran its CURTAIN CALL drone-swarm defense through a second field test at Lawrence Livermore National Laboratory in California from April 27 to May 1, 2026, and this round added the threat the program was built to stop: a coordinated attack by several drones at once, not a single aircraft.

What the command reported is narrower than the “defeat coordinated UAV attacks” framing now circulating. AFRICOM said the system detected and tracked the threats and cued a defensive swarm in near real time, with its communication and command links holding together. Then it logged where it fell short. No confirmed intercept count was released. That distinction matters, because every counter-drone program I have covered this year runs into the same wall. Tracking a swarm is the cheap part. Knocking one down reliably, at a price that beats the attacker’s, is the part nobody has fully solved.

The second test pointed the system at a coordinated attack

AFRICOM built CURTAIN CALL to throw a fast, synchronized wall of drones in front of incoming threats, and the five-day Livermore event tested that idea against both a lone drone and a coordinated multi-drone strike, the saturation tactic that has defined aerial combat from Ukraine to the Middle East.

The results, according to AFRICOM, showed the system tracking the threats and reacting faster than it did in the program’s first demonstration at the same Livermore site, which AFRICOM detailed in February. Under the concept, sensors and cameras watch the airspace, flag a possible threat, and alert a human operator who decides whether to engage. Once that operator approves, the defensive swarm launches to form an interception barrier in front of whatever is inbound. A human keeps the authority to fire.

“Our first demonstration gave us initial technical feasibility insights; this second event brought the concept to life in a controlled environment,” said U.S. Air Force Lt. Col. Jared Bindl, AFRICOM’s chief innovation officer. The team, he said, tracked coordinated threats and provided near real-time cueing of the defensive swarm, which he framed as inching the project closer to a deployable, low-cost defensive layer.

What CURTAIN CALL validated, and where it came up short

Beyond tracking, the command reported meeting a set of operational goals during the five-day test:

• Direct communication with the drone swarm, the link that lets defensive drones receive targeting data and respond as threat conditions change.
• Launching the swarm into action the moment a human operator approved the engagement, compressing the response timeline against fast-moving targets.
• Integration of the Tactical Awareness Kit (TAK), the situational-awareness software used across U.S. and partner forces, which gave operators a visual picture of the threat and shared it across command levels.

AFRICOM was also direct about the limits. Engineers used the event to find technical gaps, not only to claim wins, and the command said the data will feed its modeling and simulation work and shape the next phase of concept development. The honest read: this was an integration and tracking milestone, not a demonstration of confirmed kills against a swarm. The system showed its parts can work together. Whether it can reliably bring down a coordinated attack, and do it cheaply, is the question the next round has to answer.

The economics are the reason the concept exists

CURTAIN CALL exists because the math of shooting down drones has stopped working: defenders keep firing expensive interceptors at cheap, mass-produced aircraft, and the cost gap favors the attacker every time. A coordinated swarm multiplies that problem, forcing defenders to pay the premium again and again in the span of seconds.

DroneXL has tracked that imbalance for months. An analysis we ran in October laid out the spread: an $80 cannon round at one end and a $4.2 million Patriot PAC-3 at the other, fired at targets that can cost a few thousand dollars. Autonomous interceptor drones have started to bend that curve, with reported engagement costs around $10,000 to $15,000 and Ukrainian Sting interceptors coming in between $2,000 and $4,000, figures I walked through in covering Honeywell’s airborne counter-drone work. In May the Pentagon put real money behind the cheaper approach, a $500 million contract for combat-proven interceptors that I covered when it landed.

CURTAIN CALL’s bet runs in a different direction from a one-shot interceptor. Instead of one drone chasing one threat, it fields a swarm assembled from commercially available drones and sensors, developed with the Joint Staff’s Warfighter Laboratory Incentive Fund, so the hardware is cheap and replaceable by design. Defense officials told Stars and Stripes the effort currently flies as many as 25 counter-drones, with the potential to scale into the hundreds. That scaling logic is the same one Ukrainian startup Swarmer has been chasing, moving from 25-drone swarms toward 100 and beyond.

AFRICOM is preparing for a threat that is already on the battlefield

The command’s focus is force protection in its own theater, where forward bases and dismounted units often operate without the heavy air-defense systems that ring larger installations, and where low-cost drones have already turned up in the hands of state and non-state actors alike.

The wider picture is grimmer than a test range. A first-person-view drone struck a U.S. Army Black Hawk over Baghdad in late March, and an Iranian one-way attack drone killed six American soldiers at a site in Kuwait on March 1, casualties I detailed in covering the counter-drone funding debate. Coordinated swarms raise the difficulty further by packing the sky with targets and shrinking the seconds defenders have to react.

CURTAIN CALL also fits a wider Pentagon turn toward autonomy in threat detection and response, even with a human still holding the authority to fire. The same shift runs through the war-game scenarios the New York Times warned about late last year, and through the budget lines: the department’s counter-drone task force, JIATF 401, saw its research funding jump sharply in the fiscal 2027 request. The money is moving because the threat stopped being theoretical.

DroneXL’s Take

Here is what reads as real progress and what reads as a press release doing its job. Tracking a coordinated attack and cueing a defensive swarm in near real time is a legitimate step, and AFRICOM earns credit for publishing its gaps instead of burying them. But the verb circulating around this test, “defeat,” is doing work the command’s own statement does not. AFRICOM reported tracking and cueing. It did not report how many incoming drones the swarm actually stopped, or at what cost per engagement. Those two numbers are the entire ballgame.

In nine years of covering drone incidents over sensitive sites, I have watched the gap between what attackers can demonstrate and what defenders can reliably stop widen every year. A swarm-against-swarm answer is one of the few ideas that even tries to close that gap on the cost side, rather than on capability alone. So I want it to work. That is exactly why the missing numbers bother me.

Two things I will be watching, both grounded in what AFRICOM and its officials have actually said. First, the intercept question this test left open. The command called Livermore a controlled environment and said the next step is moving the system into a more operational setting inside its area of responsibility. Whether CURTAIN CALL holds up once the range conditions go away, and whether it posts a real intercept rate against a coordinated attack, is the open question here, not a prediction.

Second, the cost line. If a swarm of attritable, off-the-shelf drones can defend a base for something close to the $2,000-to-$4,000 per-engagement range the cheapest interceptors are now hitting, the force-protection math changes for every forward unit that currently has no good answer. If each engagement still runs into real money, this stays a science project with a good demo reel. AFRICOM has not released that number. Until it does, the verb to use is “tracked,” not “defeated.”

Sources: U.S. Africa Command (DVIDS), Stars and Stripes.

DroneXL uses automated tools to support research and source retrieval. All reporting and editorial perspectives are by Haye Kesteloo.