Space Debris Hit Seven Satellites in 2025. No Insurance Will Cover It.

Hugh Lewis was eating breakfast in Southampton when the telemetry alert arrived. It was 6:47 a.m. on January 14, 2025, and a commercial imaging satellite belonging to Planet Labs had just gone silent over the Indian Ocean. Lewis, an orbital debris researcher at the University of Southampton, pulled up the tracking data. The satellite had been healthy twelve hours earlier. Now it was tumbling, shedding fragments, its solar panels sheared off. Within six hours, the U.S. Space Force confirmed what Lewis suspected: a collision. A paint fleck—0.3 millimeters across, traveling at 15,000 miles per hour—had punched through the satellite's chassis like a bullet through tissue paper.

It was the first confirmed debris strike of 2025. By December, there would be six more. And every one of them raised the same question: who pays? The answer, in every case, was the same. No one. Because in the legal framework governing outer space, there is no mechanism to assign liability for debris strikes, no insurance market willing to cover the risk, and no treaty that can compel a satellite operator—public or private—to clean up the fragments they leave behind.

The thing is, we are now launching satellites faster than we can track them. In 2025 alone, commercial operators placed 4,312 satellites into low Earth orbit, according to data compiled by the Union of Concerned Scientists. That is more than the total number of satellites launched in the entire decade of the 1990s. Starlink, Amazon's Project Kuiper, China's GuoWang constellation, and OneWeb are building broadband networks that will eventually number 65,000 active satellites. But the Outer Space Treaty of 1967—the foundational document of space law—was written when twelve nations had launch capability and the Apollo program was the pinnacle of human ambition. It contains no provisions for traffic management, no binding rules for debris mitigation, and no court with jurisdiction to settle disputes between private companies in orbit.

What the Treaty Does Not Say

The Outer Space Treaty, signed by 114 nations, establishes that space is the "province of all mankind" and that nations bear responsibility for their national activities in space—including those of private companies. But Article VI, the liability clause, was designed for state actors launching rockets, not venture-backed constellations deploying thousands of satellites per year. When a Chinese anti-satellite test destroyed the Fengyun-1C weather satellite in 2007, it created 3,428 trackable fragments. The treaty offered no mechanism to fine China, remove the debris, or prevent future tests. When India conducted its own anti-satellite test in 2019, destroying a target satellite at 186 miles altitude, it added another 400 fragments to orbit. The international community issued statements of concern. That was all.

Brian Weeden, a former U.S. Air Force orbital analyst now at the Secure World Foundation, has spent two decades trying to build consensus on debris mitigation guidelines. The effort has been glacial. In 2007, the United Nations Committee on the Peaceful Uses of Outer Space published voluntary debris mitigation guidelines. Operators were encouraged to deorbit satellites within 25 years of end-of-life. But "encouraged" is not "required." Of the 4,312 satellites launched in 2025, only 61 percent had propulsion systems capable of controlled deorbit, according to data from the European Space Agency. The rest will tumble through orbit for decades, eventually fragmenting from thermal stress, micrometeorite impacts, or collisions with other debris.

The Insurance Market That Walked Away

Lewis, the Southampton researcher, has modeled the collision risk for mega-constellations using publicly available orbital data. His 2024 paper, published in the journal Acta Astronautica, calculated that a fully deployed Starlink network of 42,000 satellites would face a 91 percent probability of at least one catastrophic collision per year—a collision energetic enough to fragment both objects and create cascading debris. When he presented the findings to insurers at a space industry conference in London, their response was unequivocal: they would not underwrite collision risk for constellations above 1,000 satellites.

The result is a market failure. SpaceX self-insures its Starlink network. Amazon has not disclosed its insurance strategy for Project Kuiper. Chinese state-owned operators do not publish financial filings. When a debris strike disables a satellite, the operator absorbs the loss—or, in the case of state actors, the public does. But here is what this means: there is no price signal, no actuarial penalty, that discourages risky orbital behavior. A company can launch a satellite with no deorbit capability, let it drift for 50 years, and face no financial consequence when it collides with someone else's asset.

The Debris No One Can See

The U.S. Space Surveillance Network, a Department of Defense radar array, can track objects larger than 10 centimeters in low Earth orbit. But the network is blind to anything smaller. Paint flecks. Aluminum oxide particles from solid rocket motors. Fragments from battery explosions. Estimates vary, but orbital debris models suggest there are 130 million objects between one millimeter and one centimeter in orbit—each one capable of mission-killing damage at orbital velocities. And we cannot see them coming.

Moriba Jah, an astrodynamicist at the University of Texas at Austin, has been advocating for what he calls "space traffic management"—a system analogous to air traffic control, where every object in orbit has a registered owner, a monitored trajectory, and enforceable rules of the road. His nonprofit, AstriaGraph, aggregates orbital data from public and commercial sources and publishes collision predictions. On April 18, 2025, AstriaGraph flagged a close approach between a defunct Russian Cosmos satellite and an active Starlink unit. Miss distance: 12 meters. SpaceX maneuvered the Starlink satellite four hours before closest approach. The Russian satellite, dead since 2009, drifted through the conjunction uncontrolled.

Jah's data shows that Starlink satellites performed 24,000 collision-avoidance maneuvers in 2025—an average of 66 per day. Each maneuver burns propellant, shortening the satellite's operational life. And each one is voluntary. There is no international body that can compel an operator to move a satellite, even if a collision is certain.

The Legal Vacuum at 340 Miles

In January 2026, the United Nations Office for Outer Space Affairs convened a working group in Vienna to draft an update to the 1967 treaty. Thirty-eight nations sent delegates. The sticking points emerged within hours. China and Russia argued that any debris mitigation rules must apply equally to military and commercial satellites—a position the United States rejected, citing national security. The European Space Agency proposed a binding 10-year deorbit requirement for all new satellites. India countered that such a rule would disadvantage developing nations without the technology to comply. After five days, the working group adjourned with a commitment to reconvene in 2027.

Meanwhile, private operators are writing their own rules. SpaceX has committed to deorbiting Starlink satellites within five years of end-of-life—a standard stricter than the UN's 25-year guideline. But compliance is voluntary, and there is no public audit of whether satellites actually deorbit on schedule. Amazon has pledged similar standards for Project Kuiper, but the first operational satellites are not yet in orbit. China's GuoWang constellation, which will eventually number 13,000 satellites, operates under opacity. The China National Space Administration does not publish debris mitigation plans or collision data.

The Kessler Syndrome Question

In 1978, NASA scientist Donald Kessler proposed a scenario that now bears his name: the Kessler Syndrome. If debris density in low Earth orbit crosses a critical threshold, collisions will begin to generate debris faster than atmospheric drag can remove it. Each collision creates fragments. Those fragments cause more collisions. The process becomes self-sustaining, a cascade that renders entire orbital shells unusable for generations.

The scientific debate is not whether the Kessler Syndrome is possible—it is. The question is whether we have already crossed the threshold. Lewis, the Southampton researcher, argues that certain orbital shells—particularly the 500 to 600 mile band where old Russian satellites congregate—are already in runaway debris growth. His models show that even if we stopped launching satellites tomorrow, collisions in that band would continue for 200 years, each one adding fragments to the population. Other researchers, including Jah, are more cautious. They argue that active debris removal—missions that capture and deorbit dead satellites—can still stabilize the environment, if we act soon.

But here is what makes the debate urgent: active debris removal is expensive, unproven at scale, and legally ambiguous. Who has the right to capture another nation's satellite, even a dead one? Under the Outer Space Treaty, a satellite remains the property of its launching nation forever. In 2025, a Swiss startup called ClearSpace planned to test a debris removal mission by capturing a defunct European Space Agency satellite. The mission was postponed when legal advisors could not confirm that removing the satellite without explicit ESA consent would comply with international law.

What We Still Don't Know

On the morning of the first debris strike in January 2025, Hugh Lewis ran his orbital propagation software and calculated how many fragments the collision had likely created. The answer: between 40 and 300, depending on the impact angle and energy. Most would be too small for ground radar to track. They would remain in orbit for years, invisible threats to every satellite in that altitude band. Lewis filed his report with the Inter-Agency Space Debris Coordination Committee, the body that advises the UN on debris issues. The committee thanked him for the data. No action was taken.

The thing we still do not know is this: how much debris can the orbital environment absorb before the cascade begins? The models disagree. The data is incomplete. And the treaty that was supposed to prevent this scenario is fifty-nine years old, written in an era when space was the province of superpowers, not venture capital. We are conducting an experiment in orbital commons management at a scale humanity has never attempted. And unlike most scientific experiments, this one has no control group, no way to rewind the clock if the results go wrong. The fragments are already up there. We just cannot see them yet.