In 2022, a senior leader told our team the backup plan for a small drone attack on a Navy installation was buckshot. I spent the next week writing a whitepaper on why that would get people killed.
The engagement envelope was wrong. The collateral risk was unacceptable. And the threat was evolving faster than a 12-gauge could solve.
I wasn't at DARPA making strategy. I was at the installation level — the analyst who'd have to help build the case for why we needed better options, knowing full well that the options we had weren't good enough.
The Math That Changed Everything
The reality of defense acquisition: the people closest to the problem have the least authority to change the solution. CNIC deploys what it's given. PEOs decide what gets built. The PPBE cycle takes 2+ years to fund anything new.
Meanwhile, Ukraine rewrote the math.
As of January 2026, President Zelensky reported that Russia had launched nearly 29,000 Shahed-type one-way attack drones at Ukraine — at roughly $20,000-$50,000 each. A single Patriot PAC-3 interceptor costs an estimated $3.73 million (per CSIS, Feb 2025). Fire a Patriot at a Shahed and you've spent roughly 107x what your enemy spent. That's not defense. That's attrition — and you're the one being attrited.
Ukraine went from 7 drone manufacturers before the invasion to over 500 today, producing roughly 4 million drones per year. Fifteen companies now build fiber-optic FPV drones that are immune to electronic warfare jamming. Over 350,000 fiber-optic drones have been delivered to the armed forces since mid-2025.
Three Years Later: Still Buckshot
Three years after that whitepaper, in July 2024, the Marine Corps published a formal sources sought notice on SAM.gov requesting "buckshot-like" canister rounds for rifles chambered in 5.56mm and 7.62mm. They also wanted rifle-mounted RF jammers and advanced tracking optics.
By December 2025, Marines at Camp Pendleton were training with M1014 shotguns and frangible rounds against drones during Steel Knight 25.
The problem I wrote about hasn't gone away. It's gotten worse.
The Directed Energy Promise
The good news: directed energy is finally arriving. Epirus delivered four Leonidas high-power microwave systems to the Army, and in August 2025 it disabled 61 out of 61 drones — including a 49-drone swarm — with a single pulse. 100% success rate. In December 2025, it even demonstrated the ability to defeat fiber-optic controlled drones.
AeroVironment delivered JLTV-mounted LOCUST laser weapons to the Army in December 2025. Israel's Iron Beam reportedly operates at roughly $3.50 per engagement (per Missile Defense Advocacy Alliance estimates).
From an estimated $3.73 million per intercept to single-digit dollars. That's the kind of cost curve we need — if it can be fielded at scale.
But here's the problem: these systems exist as prototypes and limited fieldings. The Army's Enduring High Energy Laser program — its first program of record for counter-drone lasers — won't even begin competitive source selection until mid-2026.
The gap isn't technology — we have counter-UAS systems. The gap is speed. Speed to field them, fund them, and iterate when the threat changes faster than our budget cycle.
The Speed Gap
What Ukraine built in months, our acquisition system can't deliver in years. That's not a technology problem. That's a process problem. The same pattern played out with Replicator — where we promised thousands and delivered hundreds.
Ukraine's drone industrial base went from hobbyist workshops to 500+ manufacturers producing 4 million units annually. They did this under active bombardment, with supply chain constraints, and without the world's largest defense budget.
We have every advantage they don't — money, infrastructure, industrial base, peacetime to prepare — and we're still training Marines to shoot drones with shotguns.
Buckshot was never a solution. It was a confession — an admission that we didn't have anything better. Four years later, directed energy and electronic warfare systems exist that can close this gap. But they exist as prototypes and limited fieldings, not as programs of record at scale.
The real question isn't whether buckshot works. It's whether our acquisition cycle can outlearn a drone firmware update. Because the threat doesn't wait for source selection. It iterates weekly, and it costs less than the paperwork we file to study it.
In 2022, the backup plan was buckshot. In 2026, we have better technology — but we still don't have the speed to field it. And speed is the only metric the threat respects. The companies that understand this — Anduril, Saronic, and their peers — are the ones reshaping how we build.
Somewhere right now, another analyst at another installation is writing another whitepaper about another gap. The threat won't wait for that paper to route through the system. It never has.
Sources: CSIS Cost-Effectiveness Analysis (Feb 2025), GAO Patriot/MSMD Reports, President Zelensky via Kyiv Independent (Jan 2026), Ukraine Commander-in-Chief Syrskyi (Nov 2025), Marine Corps Systems Command SAM.gov Notice (July 2024), Epirus Press Releases (Aug & Dec 2025), Missile Defense Advocacy Alliance