Ukrainian drone founder Yaroslav Azhnyuk went from pet cameras to AI-guided weapons. He and guest host Noah Smith make the case that the West is asleep at the wheel.
The interview makes the valuable point that warfare is shifting from a paradigm centered on expensive, scarce platforms toward one centered on cheap, software-enabled, attritable systems deployed at enormous scale. Autonomy dramatically increases drone effectiveness.
However, he overreaches to claim, in essence, that drones supersede most traditional military power. He understates evolving anti-drone defenses, such as electronic warfare, directed-energy systems, automated air defense, hard-kill interception, and spectrum denial. Drones face fundamental limitations from relying on weak propulsion and onboard energy systems. Drones can't project substantial firepower over long distances. Modern military aircraft operate hundreds of miles from base, at high altitude, at supersonic speeds, carrying powerful radar, with large sensor fusion capacity, while engaging multiple airborne threats simultaneously and being capable of delivering much larger payloads than drones.
The interview rightly notes that China has a major structural advantage in drone-adjacent manufacturing. However, the Pacific is not eastern Ukraine. The Pacific involves:
* thousands of miles,
* carrier groups,
* submarines,
* maritime patrol,
* anti-submarine warfare,
* tanker aircraft,
* and fleet air defense.
Small drones face:
* endurance limits,
* weather limits,
* and sensor limitations.
The bottom line is that drones are becoming an additional mass layer underneath traditional aerospace power—analogous to how infantry, artillery, armor, aircraft, submarines, and missiles all coexisted rather than fully replacing one another.
Your aircraft comparison holds on peak capability, but the guest is arguing a different axis: cost-exchange. Directed energy only wins if its cost per shot stays under the drone's, which makes this an economics problem before a physics one. Same marginal-cost question https://thesynthesisai.substack.com/p/the-bleed-rate asks about AI inference: who subsidizes whom per unit, and for how long.
The interview makes the valuable point that warfare is shifting from a paradigm centered on expensive, scarce platforms toward one centered on cheap, software-enabled, attritable systems deployed at enormous scale. Autonomy dramatically increases drone effectiveness.
However, he overreaches to claim, in essence, that drones supersede most traditional military power. He understates evolving anti-drone defenses, such as electronic warfare, directed-energy systems, automated air defense, hard-kill interception, and spectrum denial. Drones face fundamental limitations from relying on weak propulsion and onboard energy systems. Drones can't project substantial firepower over long distances. Modern military aircraft operate hundreds of miles from base, at high altitude, at supersonic speeds, carrying powerful radar, with large sensor fusion capacity, while engaging multiple airborne threats simultaneously and being capable of delivering much larger payloads than drones.
The interview rightly notes that China has a major structural advantage in drone-adjacent manufacturing. However, the Pacific is not eastern Ukraine. The Pacific involves:
* thousands of miles,
* carrier groups,
* submarines,
* maritime patrol,
* anti-submarine warfare,
* tanker aircraft,
* and fleet air defense.
Small drones face:
* endurance limits,
* weather limits,
* and sensor limitations.
The bottom line is that drones are becoming an additional mass layer underneath traditional aerospace power—analogous to how infantry, artillery, armor, aircraft, submarines, and missiles all coexisted rather than fully replacing one another.
Your aircraft comparison holds on peak capability, but the guest is arguing a different axis: cost-exchange. Directed energy only wins if its cost per shot stays under the drone's, which makes this an economics problem before a physics one. Same marginal-cost question https://thesynthesisai.substack.com/p/the-bleed-rate asks about AI inference: who subsidizes whom per unit, and for how long.
Caitlin Kalinowski discussed this topic on Lenny's podcast recently too.