The Axis That Shares Homework: How Enemies Built a Network Without Signing a Treaty

When three American soldiers were killed at a desert outpost in Jordan in early 2024, the strike came from an Iranian-backed militia using a drone whose lineage ran straight through other countries' workshops. That is the detail worth holding onto. Russia and North Korea never fired at U.S. forces. They did not need to. Across a growing body of reporting, the pattern is the same: one member of a loose grouping — Russia, China, Iran, North Korea — supplies intelligence, components, or combat-tested technique, and another member uses it. The firing is outsourced; the learning is shared.

Be clear about what is established and what is inference. Documented: Iran sold Russia Shahed attack drones and helped Russia build them domestically; North Korean troops have fought alongside Russian forces; Chinese microelectronics turn up in Russian and Iranian systems. Inferred: whether these transactions add up to a coordinated bloc or just parallel deals of convenience. Hold that distinction; it decides everything.

The model: a network, not an alliance

The idea that explains this is the network effect — the principle, formalized by economists studying telephone and later internet adoption (Theodore Vail at Bell in the 1900s; later Robert Metcalfe and the economist W. Brian Arthur on increasing returns), that the value of a group rises faster than its headcount, because value lives in the connections between members, not in any single node. One phone is useless; a million are indispensable. Foreign Affairs labelled this grouping an "axis of upheaval." The label matters less than the structure: not a treaty alliance like NATO, but a web of bilateral deals that compound. What one member learns, the others can use.

The mechanism: how a battlefield becomes a shared lab

Follow one weapon through the web. Iran designs the drone and transfers production to Russia. Russia adds Chinese microelectronics, then modifies navigation, payloads, and tactics repeatedly to defeat Ukrainian air defenses — a fast cycle of battlefield learning. The improvements then travel back out: upgraded drones and Ukraine-tested tactics return toward Iran and its proxies; North Korean personnel absorb modern drone combat; Chinese components get stress-tested under real fire.

The key move: the network converts a single war into shared R&D. Each member contributes what it has — Iran the design, China the parts, Russia the combat testing, North Korea the manufacturing scale — and each draws out more than it put in. This is the network effect celebrated in Silicon Valley, running for America's adversaries.

The twist: the stronger network is the one decaying

Here is what cuts against intuition. The United States entered the 2020s with a vastly superior network: dozens of treaty allies, bases on three continents, interoperable forces, decades of intelligence-sharing. On paper, no contest.

Yet a network's strength is not its node count — it is how willingly nodes connect. A hundred reluctant partners can deliver less than four committed ones, because a connection nobody wants to use carries no value. When partners hesitate to grant base access or decline to join a campaign, the U.S. advantage erodes — not because it was fake, but because it is perishable. It decays every time a partner judges the cost of cooperating higher than the benefit. The adversary web, thinner but eager, moves the other way.

The other view, taken seriously

The strongest counter-reading: this is overstated. These four are opportunists with clashing interests, not a bloc. Russia and China carry centuries of mutual distrust; Iran and North Korea offer China little it genuinely needs. On this view the "axis" is an analytical convenience that dissolves the moment the deals stop paying.

This is fair, and the honest verdict is mixed. The transfers are documented; their permanence is not. What binds the four is not loyalty but a shared objection — each fears American power to isolate, sanction, and threaten its regime. As long as that fear holds, the deals keep making sense despite the suspicion. If it fades, the web frays. That is precisely the variable to watch.

Where else this shows up

The same principle runs open-source software: no firm owns Linux, yet it powers most of the world's servers, because each contributor's fix becomes every user's gain. It runs research consortia, franchises, and criminal syndicates alike — structures that win not by central command but by letting nodes teach each other. And it runs your working life: a team of brilliant people who hoard knowledge is weaker than an average team that shares it. The asset is not the talent in the room; it is the willingness to connect it. Wherever you see value pooling in the links rather than the parts, expect the eager network to beat the larger reluctant one.

What to watch, and a testable call

Two signals separate the two readings. First, whether Ukraine-developed drone upgrades — harder-to-intercept, jet-powered variants — appear in attacks on U.S. or allied forces elsewhere. That is the web circulating improvements as described. Second, whether U.S. partners drift back toward Washington or keep their distance; sustained drift is the American network eroding in real time.

A falsifiable expectation: roughly 60% that within two years, weapons or tactics clearly traceable to the Ukraine war appear in an attack on U.S. or allied forces in another theater. If the four are a durable system, this happens; if they are mere opportunists, the circulation stalls. Watch which way it breaks.

Distilled from Foreign Affairs