Preprint proposes using inspector satellites to detect hidden nuclear warheads in constellations

A newly posted arXiv preprint proposes a concrete way to check whether large satellite constellations are hiding nuclear warheads in orbit, offering a simulation-backed concept for verifying one of space law’s oldest bans. The paper suggests that, under certain assumptions, tens of inspector spacecraft could examine an entire constellation over several years and detect a single deployed warhead with high probability.

The study is a preprint, not a peer-reviewed paper. It was written by Patrick Huber of the Center for Neutrino Physics in the Department of Physics at Virginia Tech. In plain terms, the proposal would send inspector spacecraft close to target satellites, fire a high-energy 16 MeV photon beam at them, and look for the telltale neutrons produced if the beam triggered fission in hidden nuclear material. The paper says the beam could be produced with laser-wakefield accelerator and inverse-Compton methods, while the neutrons would be picked up by segmented organic scintillator detectors. As Huber writes in the abstract, “We present an analysis of active interrogation in space with the goal to verify the absence of nuclear weapons hidden in a large constellation of satellites.”

To test the idea, Huber used SpaceX’s Starlink as a real-world example constellation, modeling inspections of 7,852 satellites selected from March 2026 orbital catalog data. That is the paper’s dataset, not an allegation about Starlink. The mission scenario spans five years and combines simulations of nuclear interactions with orbital routing and maneuvers. The paper says it used GEANT4 for the radiation and signal modeling and Orekit for the orbital analysis. Its headline result is that, depending on assumptions about the accelerator system, between 20 and 70 inspection platforms could inspect the full modeled constellation with a 90% probability of detecting a single deployed warhead.

The paper also sketches what those numbers might look like in a more specific design. One architecture says inspecting all 7,852 modeled satellites within five years at a 90% probability of detection would require 69 co-orbiting inspector platforms. A smaller effort would produce weaker results: the paper says a 10% random sample of the constellation, or 785 satellites, could be inspected by two to 20 platforms, yielding a 37% detection probability against five deployed warheads.

Just as important as the headline numbers are the conditions attached to them. The paper explicitly assumes inspections would happen in a “cooperative framework under a negotiated agreement” because the method requires both close approaches and deliberate radiation exposure of another satellite. Huber reports that no inspected satellite in the modeled campaign would receive more than 0.25 gray of radiation dose, which the paper says is less than 1% of the 20 to 100 gray a typical satellite might accumulate from the natural space-radiation environment over five years. Even so, the concept would depend on international consent and rules that do not currently exist.

That gap is why the proposal matters. Article IV of the 1967 Outer Space Treaty states: “States Parties to the Treaty undertake not to place in orbit around the earth any objects carrying nuclear weapons or any other kinds of weapons of mass destruction, install such weapons on celestial bodies, or station such weapons in outer space in any other manner.” But while the treaty bans orbital nuclear weapons, it does not create a standing verification body or spell out how states would inspect spacecraft to prove compliance.

The preprint does not describe an existing international program or any state-approved inspection system, and the research report found no evidence that any country has proposed or accepted an active-interrogation regime of this kind. It also relies on technology that is not yet operational in space. Compact laser-wakefield and inverse-Compton sources have been demonstrated in laboratories and remain an active area of research, but turning them into autonomous, space-qualified inspection platforms would be a major engineering step.

Interest in that verification problem has grown since 2024, when U.S. officials publicly raised concerns that Russia might be considering nuclear weapons in its counterspace programs. That backdrop helps explain why ideas like Huber’s are resurfacing. But the paper is best read as a technical proposal for closing a treaty-enforcement gap, not as evidence that such an inspection system exists today or is close to deployment.