As Ukraine deploys $2,100 FPV interceptors to down low-altitude Russian munitions, the Pentagon is pivoting toward low-cost, AI-driven defenses to bridge the massive spending gap against swarming threats.
By: Tuva Siegel, Warrior Editorial Fellow
The Patriot PAC-3 MSE rings in at an approximate cost of 3 to million, 60 to 100 times the cost of the threat, the Iranian Shahed-class One-Way Attack Drone. Ukraine has responded to that gap by fielding FPV interceptors that engage the same threat for as little as $2,100. The Pentagon is paying attention.
According to Counter-Drone 101, an April 2026 reference document by Vice President of Raincloud Defense, George Howell. “The Shahed and its derivatives are long-range, low-flying, propeller-driven loitering munitions designed to fly under traditional air defense coverage.” Previously, their success was demonstrated in Saudi Aramco's Abqaiq and Khurais facilities in 2019, destroying the radar gap from medium to high-altitude defense systems, as opposed to more easily detected larger medium-altitude, non-stealthy drones. Shaheds are small, fast, and low-altitude drones. Lower altitude increases the possibility that a drone could fly beneath the radar aperture and be more difficult to detect.
The Shahed remains a threat “that most counter-drone procurement is currently scoped against,” with a price range of $20–50K, significantly less than the U.S. Patriot. As Howell explains, “The widely-cited framing of a Patriot interceptor at several million dollars destroying a $30,000 Shahed is not really how the engagement is being fought.” In other words, the old logic that frames bigger, heavier, and more expensive as the winner is an old logic and is now dead in the face of Shahed and its variants.
The Iranian-made Shahed-136 has been developed into a variety of different variants, mainly Russian, such as the Geran-2, costing approximately 80k per unit, upgrading the Iranian models to long-range OWA. 170+ per day production and ability to fly under medium/high- altitude SAM coverage to include AI-targeting and GPS-denied nav. “ISIS data shows Russia launched 6,462 Shahed-type UAVs in March 2026 alone, averaging 208 per day with a single 24-hour peak of 942 launches.” However, the progress does not halt here. Even the one-way attack (OWA) Shahed must meet its superior, which closely aligns with what Howell outlines as the most necessary considerations from drone capabilities: “aligned with where the threat is going (acoustic, mmWave with micro-Doppler, visual AI, edge inference, GPS-denied architectures)”, these standards and features have developed from none other than Ukraine, “the centre of operational learning.”
Operation Epic Fury has been heavily aligned with Ukraine’s best, having “absorbed 1,422 detected drones plus 246 missiles in the first week (CSIS); Bahrain intercepted 70 missiles and 59 drones by 3 March; Gulf states reported 80–90 percent interception rates,” Howell states.
In a February 6, 2026, U.S. Army article detailing JIATF-401’s acquisition of the Bumblebee V2 counter-drone system, Army Lt. Col. Adam Scher explains, “The Bumblebee V2 is a next-generation FPV multirotor drone. The system is designed to physically intercept and neutralize hostile small unmanned aircraft systems (sUAS). This is achieved through a drone-on-drone collision when the Bumblebee V2 directly engages the threat, rendering both aircraft inoperable.” The $5.2 million agreement with Perennial Autonomy of the Bumblebee aligns with “the JIOP's goal to accelerate the delivery of top-tier technology to warfighters.”
In its own war, Ukraine’s Wild Hornets' Sting first-person-view (FPV) interceptor accounts for around 70% or even 100% on certain days, according to the manufacturer's website, making it the most effective interceptor against Shahed-type drones in 2026 at only $2,100–$2,500. Even less costly is the Fiber-optic 3D-printed P1-SUN (SkyFall, UA) at only $1,000–$3,000, with over 3,000 Shahed kills in 2026. These two drones are small Unmanned Aerial Systems, “cheap, agile, often first-person-view piloted, and increasingly autonomous.”
The Counter-drone strategy and the rise of its industry are at the core of Howell’s argument, as its defense procurement and deployment context depends heavily on several factors as “A forward operating base, a vehicle convoy, a fixed installation with grid power, an aircraft carrier, a helicopter, and a stadium each demand a different architecture” will determine whether “sensors, the effectors, the rules of engagement, and the C2 posture all change.” Ukraine’s battlefield experimentation and urgency have propelled this service into existence; the counter-drone as an ongoing service rather than a capital purchase is an appealing step into the future. “The Pentagon's FY2026 baseline request for counter-drone across the services is $3.1 billion.” The FY2026 Budget Request Overview adds that terminating future investments such as “the M10 Booker, High Mobility Multi-Wheeled Vehicle (HMMWV), Joint Light Tactical Vehicle (JLTV), Robotic Combat Vehicle (RCV), and the MQ-1C Gray Eagle” will improve and diversify “unit survivability and mobility, as well as free up resources to reinvest in cutting-edge technologies that more directly enhance warfighting capability and readiness”
However, destroying drones is only part of the challenge; finding them has become equally important. The rise of fibre-optic drones has exposed major weaknesses in counter-drone systems built around radio-frequency detection and GPS jamming. Unlike conventional FPVs, fibre-optic drones emit no RF signal, so jamming and spoofing have no effect.” Their effectiveness was demonstrated during the Pentagon's Operation Clear Horizon exercise in 2025, where “special operators flew fibre-optic and LTE-controlled drones (with pilots in Colorado striking targets in Florida) against US counter-drone defences and exposed the gap directly.” Howell argues that systems relying on assumptions “that the threat will be RF-controlled” already have “a depreciation curve baked in.” Instead, capabilities “aligned with where the threat is going,” such as acoustic sensing, millimetre-wave radar, visual AI, and GPS-denied architectures, are becoming increasingly valuable.
Ukraine's Sky Fortress and Bullfrog are acoustic and visual AI networks (respectively), with the Sky Fortress consisting of thousands of low-cost sensors integrated into a national air picture that reportedly detects around “20 percent” of aerial targets through acoustic means alone. As drone warfare evolves and warfare must constantly “look up”, the challenge is no longer simply intercepting the threat but accurately detecting and identifying it before an interceptor is ever launched. “Ukraine's Sky Sentinel autonomous .50-cal turret and the C2 stacks behind them are all attempts to algorithmically execute the kill chain at speeds humans cannot match,” says Howell.
Recently, Ukraine's Sky Sentinel has been investigated by the U.S. as an alternative to the expensive Patriot drone. The turret can “track and engage fast-moving targets, traveling at speeds between 200 and 800 kilometers per hour. A substantial accomplishment for even the U.S.’s SHORAD systems, such as the Stryker M-SHORAD Increment I or the Counterrocket, Artillery, and Mortar (C-RAM) Weapon System, to achieve,” explains a center for Army Lessons Learned paper examining Ukraine’s AI-powered Sky Sentinel air defense turret and its effectiveness to inform U.S. Army SHORAD and C-UAS development, including AI integration and production decisions. The Sky Sentinel uses AI to calculate drone speed, adjust for wind resistance, and predict the exact point where the bullet and drone will collide, even managing to differentiate between birds and small UASs, while being semi-autonomous, but with human authorization still needed. However, like with anything, “factors such as adverse weather conditions, electronic warfare (EW), complex urban environments, and the presence of very small or fast drones can significantly diminish the effectiveness of the Sky Sentinel.” It is estimated to cost about $150,000 per turret. This is plausible because Sky Sentinel combines a heavy machine gun (such as a Browning M2 .50 caliber), a stabilized mount, COTS radar and electro-optical sensors, and custom Ukrainian AI fire-control software. Rather than relying solely on expensive foreign systems such as Patriot, Ukraine prioritizes scalable, domestically developed solutions.
Tuva Siegel is an Editorial Fellow at Warrior Maven. She studies English at Kenyon College. Tuva is the author of Drömland, a fictional collection of short stories, and is currently studying weapons and military technology.