Nobody can currently prove whether a nuclear weapon is hidden in orbit

The 1967 Outer Space Treaty bans any country from placing nuclear weapons in orbit, but the agreement has never included a way to check whether a satellite is actually complying. No government has an approved, unclassified method for peering inside a suspect spacecraft to confirm or rule out a hidden warhead. An MIT physicist has now published a peer-reviewed proposal for how such an inspection could work, but it remains a feasibility study on paper, not a system that has ever been built or flown.

A 1967 promise with no way to check it

The Outer Space Treaty, which entered into force in October 1967, commits its parties not to place nuclear weapons or other weapons of mass destruction in orbit around Earth, on the moon, or elsewhere in space. What the treaty’s text does not provide is a verification mechanism. Article IV bars states-parties from placing nuclear weapons or other weapons of mass destruction in orbit, installing them on celestial bodies, or stationing them in space by any other means, but it leaves enforcement to trust rather than inspection. There is no inspection regime, no agreed sensor standard, and until recently no peer-reviewed method proposed anywhere in the unclassified scientific literature for confirming what a given satellite is actually carrying.

Why a detonation in orbit would be so damaging

The concern is not hypothetical history. In 1962, the United States detonated a 1.4-megaton thermonuclear device in space, a test that inadvertently destroyed a number of the era’s satellites by flooding the Van Allen radiation belt with high-energy electrons stripped from the bomb itself. Those electrons became trapped in Earth’s magnetic field, where Danagoulian says they continued to damage spacecraft electronics passing through the belt long after the initial blast. A modern detonation in low-Earth orbit, the band roughly 100 to 1,200 miles up where most satellites operate, would do the same thing at a much larger scale, disabling communications satellites, GPS constellations and space-based internet networks across multiple countries at once, according to MIT nuclear science and engineering professor Areg Danagoulian.

The satellite that renewed the question

Concern about the enforceability of the treaty — which entered into force in October 1967 and has been joined by well over 100 countries including the United States, Russia and China — sharpened after Russia launched a satellite called Cosmos-2553 into an unusually radiation-heavy orbit in 2022, a zone most operators avoid because of the damage it does to onboard electronics. Russia has said the satellite is used for surveillance and sensing. U.S. officials have said they suspect it may carry components of a nuclear device under testing, though that assessment has not been independently verified, and no government has presented public evidence that a weapon is currently in orbit. The episode is what prompted Danagoulian to ask whether the underlying scientific question, proving what a satellite contains without opening it, could be answered at all.

Reading a satellite’s contents from a distance

Danagoulian’s proposal, published as a peer-reviewed paper in Nature, centers on a physical process called spallation. When a high-energy proton strikes a heavy radioactive element such as uranium or plutonium, it can knock loose dozens of neutrons in a single collision, millions of times a second under the right conditions. Ordinary satellite components would not produce anywhere near that neutron signature. Danagoulian’s design pairs neutron-sensing scintillator panels with synthetic diamond detectors that filter out the naturally occurring protons and electrons already present in low-Earth orbit, allowing the system to isolate neutrons coming specifically from a nearby target satellite and estimate the direction they came from.

What the numbers say the system could actually do

By Danagoulian’s calculations, a detector package roughly the size of a large encyclopedia could identify a hidden nuclear weapon with more than 99 percent confidence if it orbited within about 4,000 meters, roughly 2.5 miles, of a suspect satellite for around a week. Closing that distance would sharply cut the time needed: at 1,000 meters, he estimates the same detection confidence could be reached in about an hour, effectively a single close flyby. Those figures come from modeling, not from an inspector satellite that has actually flown a mission, and Danagoulian describes the paper explicitly as a feasibility study rather than proof that a working system exists.

Still a proposal, not a deployed capability

Danagoulian says the immediate goal is narrower than fielding hardware: he wants national laboratories to build on the physics and policymakers to weigh the concept as a future verification tool, not to claim the problem is solved. He is also working with MIT’s Center for Nuclear Security and Policy to think through how such a system would fit into arms-control policy if it were ever built. The research was supported in part by the National Nuclear Security Administration, along with the Carnegie Foundation and Longview Philanthropy, reflecting the mix of government and nonproliferation-focused funders interested in the question. His own summary of the stakes is blunt: intelligence claims about a hidden weapon can be disputed, he says, but a neutron count from a working detector cannot be argued with in the same way. For now, though, no such detector exists in orbit, and the honest answer to whether any country currently has a nuclear weapon hidden in space remains that nobody, including the treaty’s own signatories, has a verified way to know.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview