Drone Warfare Strategy

Dealing with revolutionary weapons

Noah Smith writes,

My interview with Azhnyuk clarified exactly why drones are in the ascendant as the universal modern weapon of war. The reason is cost. Drones are simply so cheap to produce in huge numbers that they can overwhelm any more expensive system.

Mackenzie Eaglen writes,

A recent bipartisan effort to harness battlefield lessons from Ukraine on drone warfare by Senators McConnell and Bennet cannot go fast enough.

…According to Steve Blank at Stanford, the Earth’s surface is contested space, and “undefended high value fixed civilian infrastructure” is all at great risk. More simply: “the long-term math no longer favors the defender.” Placing more military assets underground can “preserve forces and enable maneuver.”

For the WSJ, Walter Russell Mead writes,

The fundamentals of drone warfare favor the defense over the offense, meaning that Ukraine’s smaller armed forces can hold off larger numbers of Russian attackers while extorting a high price for every acre of ground gained. That could change overnight, however, and both sides are scrambling in a conflict unfolding amid a revolution in military technology.

If the defensive goal is to repel an invader, then as of now drones create a big advantage to the defender. An invasion requires large, visible forces. These can be mauled by low-cost drones.

If the defensive goal is to protect civilian infrastructure and military bases, then drones give the attacker the advantage. As a defender, you have to bury and/or disguise your important assets.

In the two years since I last offered amateur speculations on this topic, drone warfare has advanced farther. As of now, there seems to be no good defense against a drone swarm. Some drones are going to get through, and they will destroy humans and expensive military equipment. The cost-effectiveness of drones means that military conflicts will be decided by which side deploys drones more successfully.

Without using any ships, Iran is asserting control over the Strait of Hormuz. The threat posed by drones and missiles is sufficient to deter us from risking using expensive ships to patrol the waterway.

Blinding Glimpses of the Obvious

I am reading Beast in the Machine, a treatise on robotic warfare by George M. Dougherty. Much of what Dougherty writes strikes me as Blinding Glimpses of the Obvious. But it may be important for his message that drones are a revolution gets through to a military that is bureaucratic and resistant to change.

Although it was published late in 2025, it seems to have been composed a few years prior. The index has only two entries for “Gaza,” neither of which refers to the war that Hamas launched in October of 2023.

Dougherty mentions large language models briefly and dismissively.

deep learning works best in orderly, easily digitized, and carefully bounded virtual environments.

In contrast to what deep-learning AI requires, war is perhaps the most volatile, uncertain, complex and ambiguous of all environments. p. 143

People who claim that their domain of expertise is unsuitable for AI usually turn out to be wrong. I hope that America’s military leaders are not taking Dougherty too seriously here.

But I agree with Dougherty when he writes

On the future battlefield ruled by precision weapons, anything that can be seen can be hit and killed. Therefore, we can expect future forces to strive not to be seen, while making maximum effort to locate the enemy. Combat may change from a struggle to hit the enemy into a struggle to find and target the enemy. p. 83

My bet is that AI will be very important as a tool for finding the enemy.

As Dougherty points out, the lethality of small machines reduces the relative value of bulky, armored weapons systems. It raises the relative value of weapons that can be easily hidden, dispersed, and replaced at low cost.

In the WSJ, Tom Tugendhat writes,

Since February, the U.S. has fired more than 1,300 Patriot interceptors against Iranian missiles and drones. Each interceptor costs around $4 million to destroy weapons that cost between $20,000 and $50,000. Based on the most recent rate of production, it will take two years for Lockheed Martin to replace what has been fired in the past 2½ months. That is the economics of defeat, and our adversaries understand it.

I have no military experience or qualifications, but as an economist I suspect that the enemy’s drones and missiles, once launched, are difficult to stop in a cost-effective manner. Instead, it seems to me that it is more efficient to do something about their weapons while they are still on the ground.

What about destroying the enemy’s drone storage facilities? A lone drone in storage is invisible. But thousands would be a target.

What about impeding the enemy’s drone production facilities? I would expect that their military strategists are putting these facilities deep under ground. But perhaps they still can be neutered.

Because drones require many components, manufacturing capacity must depend on the country’s transportation infrastructure. If the roads and rails leading to the manufacturing plants become impassable, production will become impossible.

I would be trying to pinpoint the locations of launch sites, storage sites, and manufacturing sites. Even if these sites are securely located underground, they might be rendered unusable by destroying the surrounding transportation infrastructure.

Dougherty points out how countries like Iran or Ukraine, and non-state militias like Hamas, can be effective without being able to defeat their adversaries in 20th-century-style wars. Drones and other robotic weapons level the playing field against large industrial powers.

Underdog forces are able to maintain conflicts that fall in the gray zone between war and peace., waiting for the industrial powers to tire of fighting. This confounds western societies, which despair of achieving decisive victory. We call these gray-zone conflicts “forever wars,” and we walk away from them.

In contrast, my understanding is that there is a long tradition in Muslim warfare of avoiding decisive defeats by retreating and/or offering a ceasefire. They renew fighting on their terms.

As drones achieve parity with heavy industrial weapons, delaying tactics can become more effective. If we want to win a war, we will need to find a solution.

substacks referenced above:

@

All non-drone militaries are obsolete