Ukraine’s Faster-Drone Problem Cannot Be Solved With Missiles Alone

Ukraine’s effort to build cheap defenses against Russia’s faster drones hit a serious obstacle in July: during military trials, several new interceptors became difficult to control at high speeds, and some crashed. Reuters reported those failures on September 15, alongside Ukrainian military data showing launches of drones traveling at 240–500 kilometers per hour rising from roughly 450 in June to 2,850 in August. Those counts are not independently audited, but the reported combination is alarming: a growing threat and an unfinished response. The question is how to close that gap without exhausting the weapons needed against other attacks. (internazionale.it¹)

The immediate case for buying more missiles is persuasive. Cities need protection before experimental aircraft become reliable. Established missile systems also have a substantial record: Kongsberg, which makes the NASAMS ground-based air-defense system, reported more than 900 interceptions of missiles and drones in Ukraine and a 94% success rate in 2025. That is a manufacturer’s aggregate claim, without a jet-drone-specific breakdown. Still, it deserves more weight than a prototype’s advertised maximum speed. (kongsberg.com²)

I would nevertheless prioritize expanding the network of sensors and proven, lower-cost defenses, while replenishing missiles for threats those defenses cannot reliably handle. The distinction matters: missiles must be available immediately when needed, rather than withheld until a cheaper attempt fails. But making greater missile consumption the main response would deepen an existing supply problem. Air-power researcher Justin Bronk’s assessment for the Royal United Services Institute identifies ammunition resupply as the central constraint on Ukraine’s surface-to-air missile coverage. (rusi.org³)

First, the threat needs a more careful description than “jet Shahed.” Ukraine’s military intelligence agency identifies a Geran-4 with a redesigned airframe and jet engine. Yet its public specifications differ: a May 25 announcement describes speeds up to 500 km/h, an altitude ceiling of 5,000 meters and 50- or 90-kilogram warheads; its technical portal lists 350 km/h maximum speed, a 4,000-meter ceiling and a 50-kilogram warhead. These are attributed Ukrainian assessments, not independently established performance limits, and the pages do not reconcile the differences. Neither tells us the actual altitude and speed distribution of recent attacks. (gur.gov.ua)

That uncertainty matters because maximum speed alone does not determine the engagement window, the time available to detect a threat and stop it. In a deliberately simplified calculation, an aircraft crosses 20 kilometers in six minutes at 200 km/h, but in just 2.4 minutes at 500 km/h. Those are assumed speeds and distance, not an observed interception. Identification, authorization and launch must fit inside the available time. Terrain, flight altitude and sensor coverage can change when that clock starts; the Army’s acoustic-detection assessment specifically describes coverage limits and the need to combine sensors. (api.army.mil⁶)

This makes detection a procurement priority in its own right. A U.S. Army paper published in April describes thousands of Ukrainian acoustic sensors sharing detections to locate drones and alert defenders. These passive listening devices complement radar and optical sensors; they do not replace them. The paper also describes limits from background noise and atmospheric conditions. I would fund their integration with radar, resilient communications and a common tracking picture before treating another advertised interceptor speed as a breakthrough. A warning that reaches the right crew early can be more useful than a faster aircraft launched too late. (api.army.mil⁶)

Electronic warfare, which disrupts or deceives a weapon’s navigation and communications, also belongs in that network, but should receive no presumption of success. Ukraine’s Geran-4 component inventory identifies an anti-jamming system using a 16-element antenna. That establishes equipment designed to resist interference, not immunity to every countermeasure. I would require evidence that disruption actually prevented a damaging arrival, rather than count a lost track as a saved building. (war-sanctions.gur.gov.ua)

Guns face a different constraint: coverage. Bronk describes Ukraine’s cheaper short-range defenses as effective collectively because they are deployed densely and in depth, while each protects only a small area. They also require substantial manpower. His assessment credits these defenses with most of the Geran and Gerbera drones shot down, without establishing a separate success rate against the newer jet variants. That supports retaining guns where they work, not assuming that adding more gun teams automatically provides dependable protection against every faster flight profile. (rusi.org³)

Interceptor drones, unmanned aircraft built to destroy other aircraft, offer a more scalable manufacturing path. Associated Press reporting documented models costing as little as $1,000 and a progression from prototype to mass production within months in 2025. That is meaningful industrial evidence. It does not establish that the same designs can reliably defeat the faster threats now arriving. (apnews.com⁷)

The newer claims need the same discipline. In July, SkyFall told Reuters that its P1-SUN Jetkiller could reach 370 km/h and had destroyed more than a dozen jet-powered Shaheds. Reuters could not independently verify those battlefield claims. Successful encounters also leave crucial questions unanswered: how many attempts failed, and how many targets could never be engaged? Ukraine’s air force has separately reported successful head-on interceptions, while acknowledging that further development was needed before large-scale deployment. That is a reason to accelerate testing, not book future protection as already delivered. (internazionale.it⁸)

Missile procurement deserves equally unsentimental accounting. RTX announced in February that annual production of AMRAAM, a missile used in airborne and ground-launched roles, would rise above 1,900. A related rocket-motor factory expansion is expected to become operational in 2027. Neither announcement specifies Ukraine’s allocation or monthly deliveries. One missile program cannot represent the entire defensive inventory, but the distinction between future factory capacity and ammunition available tonight is unavoidable. Promised replenishment cannot justify unlimited current consumption. (rtx.com⁹)

Nor should procurement officials confuse purchase price with cost per successful interception. The published $1,000 drone price and SkyFall’s claimed manufacturing capacity do not provide a complete operational bill. I would require contracts to account for sensors, crews, training, unsuccessful launches, maintenance and ammunition used per verified success. More importantly, I would count threatening aircraft never engaged. A system can look excellent when measured only against the targets its operators chose to attack. The available price and production reporting does not resolve that denominator. (apnews.com⁷)

Damage reporting makes clear why interception totals are insufficient. Reuters photographed rescuers carrying jet-drone wreckage recovered from a Kyiv apartment building on September 1. On September 17, AP reported authorities’ accounts of at least 19 people injured in Kyiv and its surrounding region, while Ukraine’s air force claimed 131 aerial targets destroyed or suppressed during a mixed attack. Those reports establish harm alongside claimed defensive successes; they do not isolate the damage attributable to jet drones or compare competing defenses. (reutersconnect.com¹⁰)

I would therefore judge defensive combinations by verified damage prevented and their ability to sustain that protection, using confidential operational audits where public disclosure would endanger crews. Procurement should reward actual deliveries and repeated performance against identified variants, including missed engagements, rather than selected kill videos. The public evidence reviewed here does not supply that comparison, so a precise budget split would be false confidence.

The choice I favor is demanding: preserve immediate missile coverage for otherwise unprotected critical targets, while directing expansion toward shared detection and cheaper systems as they demonstrate useful protection. To favor missile expansion as the principal long-term answer, I would need evidence of replenishment that can sustain the burden without weakening defense against other weapons. To replace missile coverage with inexpensive drones, I would need evidence of protection rather than promises of production. Ukraine cannot afford to mistake either kind of promise for a working defense.