What Ukraine Is Teaching Us About Defending a State With Drones

I was born in Ukraine. I have spent close to a decade building autonomous drones — on the civilian, infrastructure side, the dual-use side, never as a weapons company. So when I watch what is happening in my home country, I am not reading it as news. I am reading it as the clearest live experiment we have ever had in how drones change the defense of a state.

Let me be precise about what this piece is and is not. It is about defending — sensing, protecting infrastructure, stopping hostile drones, keeping critical systems running under pressure. It is not a guide to offensive systems, and I won't write one. The lessons that matter most for the rest of us are defensive anyway.

Here is what I take from it.

1. The economics of defense inverted

The single biggest lesson is about cost. A drone that costs a few hundred or a few thousand dollars can now threaten an asset worth millions. That asymmetry rewrites the defender's problem. The question is no longer "do we have a powerful enough system?" It is "what does each intercept cost us, and can we afford to do it a thousand times?" Ukraine's answer has been the interceptor drone: it produced 100,000 of them in 2025, and its National Security and Defense Council says downing an attack drone this way costs more than 25 times less than firing a Western air-defense missile.

States built around a small number of exquisite, expensive platforms are structurally on the wrong side of that math. The ones adapting fastest are building cheap, layered, autonomous defenses — many modest systems working together — rather than betting everything on a handful of irreplaceable ones.

2. The spectrum is contested, so autonomy wins

In Ukraine, electronic warfare is everywhere. GPS jamming and spoofing are routine. A drone that depends on a clean satellite fix and a constant radio link is a drone an adversary can blind or take over.

This is the lesson I feel most directly as a builder. For years, "autonomy" was a convenience feature — fly the mission without babysitting it. The war turned it into survival. If the link drops and the GPS lies, the drone has to navigate, decide, and finish the job on its own. That is why authenticated navigation matters: Galileo's OS-NMA service, operational since July 2025, makes spoofing far harder to get away with, and building a real platform around it — as we did in the Horizon 2020 HUUVER project — is no longer academic. Resilience to a contested spectrum is the whole game.

3. Counter-drone is half the picture

We spent years talking about what drones can do. Ukraine made the other half unavoidable: what do you do about the other side's drones?

Counter-UAS — detecting, tracking, and stopping hostile drones — is now a co-equal discipline. And the hard-won insight is that no single method works in every condition:

• Jamming is cheap and fast, but it does nothing against drones on fiber-optic links or fully onboard navigation, and it creates collateral spectrum problems of its own.
• Kinetic shoot-down works, but with missiles it is expensive — back to the cost-per-intercept problem, and the reason Ukraine turned to cheap interceptor drones.
• Physical capture — net interception and similar — fills the gap for protecting fixed, sensitive sites where you cannot flood the spectrum or fire freely.

A serious defense layers all three. I have seen the capture side up close: in AUDROS, a project backed by ESA and the EDA, a partner's net-capture drone was housed in our docking station and launched automatically on a cue from the detection system, and the consortium tested it in the Czech Republic in 2022.

4. Production sovereignty is a capability, not a logistics detail

Ukraine turned domestic drone production into a strategic asset — building at home, iterating in weeks, and steadily cutting its dependence on Chinese components. In 2024, 96% of the more than 1.5 million FPV drones its defence ministry bought came from Ukrainian makers. Attrition is won by whoever can do that.

The peacetime version of this lesson is already written into procurement policy. The EU's European Defence Industrial Strategy (EDIS) invites member states to spend at least half of their defence procurement budgets within the EU by 2030, and Section 848 of the US FY2020 NDAA bars the Pentagon from buying or operating drones made in China or built with Chinese flight controllers, radios or cameras. Both push buyers toward allied, auditable, non-adversarial supply chains. That is not protectionism for its own sake — it is the recognition that a supply chain running through a potential adversary is a vulnerability you cannot patch in a crisis. Sovereign capacity is now part of the defense itself.

5. The advantage is the feedback loop

The platforms get the headlines. The real edge is the speed of the loop between the field and the workshop. The side that observes, modifies, and redeploys faster wins — even with simpler hardware. Software-defined behavior, over-the-air updates, and a culture of weekly iteration beat a beautiful airframe that takes two years to revise.

I have believed this my whole career, and the war has been a brutal confirmation. Build the loop, not just the product.

6. The home-front lesson — for states not at war

If your country is not on a front line, it is tempting to file all of this under "someone else's war." That is the mistake.

The surfaces a hostile drone can surveil or disrupt are the same ones every country already worries about: power grids, rail, refineries, ports, borders, stadiums, government sites. And the autonomy stack that is best at monitoring that infrastructure is the same one best positioned to help protect it. This is the dual-use reality, and it runs in both directions. The drone-in-a-box that inspects a refinery is one configuration change away from providing persistent watch over the same site.

That is the work I have chosen — the civilian, infrastructure side of a dual-use field — and I am deliberate about the line. Dual-use is not weapons. The goal is resilience: keeping the lights on, the rails safe, and the airspace over a sensitive site under control.

What I carry out of it

I did not need the war to tell me autonomy, resilient navigation, counter-drone capture, and sovereign manufacturing matter. But it removed any doubt about the order of priorities. Build for a contested spectrum. Assume the link will drop. Make the defensive layer affordable enough to use at scale. And build it somewhere you control, with a supply chain you can stand behind.

I'm Ukrainian. I build this in Poland, for Europe and the US, as dual-use infrastructure. Watching home has made me more certain, not less, that the defensible version of this technology — the one that protects people and the systems they depend on — is the one worth building.