Starlink Controls the Battlefield. Ukraine Learned What Happens When Musk Decides.

Lieutenant Colonel Denys Sharapov was three kilometres from the Russian line when his drone feed went black. It was September 2023, near Bakhmut, and the screen that had shown enemy positions moments before now displayed only static. He checked the Starlink terminal mounted in his armoured vehicle. Connection lost. He tried again. Nothing. Around him, across the Donetsk front, hundreds of Ukrainian units were experiencing the same sudden silence.

The outage lasted eleven minutes. In that window, Ukrainian forces lost real-time intelligence on Russian artillery movements, could not coordinate drone strikes, and reverted to radio communications the enemy could intercept. When service returned, Sharapov filed a report. The reply from Kyiv's digital warfare command was terse: the terminals had worked exactly as designed. SpaceX had geofenced the area—Starlink connectivity stopped at coordinates SpaceX determined were inside contested territory, likely to prevent Ukrainian drones from striking deep into Russian-held regions.

The incident revealed something military strategists had discussed in theory but never witnessed in practice: a private American company controlled battlefield communications infrastructure for a sovereign nation at war, and could alter its function—deliberately or through corporate policy—without warning. No treaty governed this. No international body adjudicated disputes. Elon Musk, who owns SpaceX, had become a stakeholder in a European land war, whether he intended to or not.

The Constellation That Changed Warfare

Starlink was never designed for war. SpaceX began launching satellites in 2019 to provide broadband internet to underserved markets—rural Montana, the Australian outback, remote Pacific islands. The business model was straightforward: low Earth orbit satellites, cheaper than traditional geostationary systems, priced for consumers. By February 2024, SpaceX had deployed 5,874 active satellites, more than all other operators combined. The constellation covered every continent.

When Russia invaded Ukraine on February 24, 2022, the Ukrainian government's internet infrastructure became a primary target. Russian cyberattacks took down servers; missile strikes severed fibre-optic cables. Within forty-eight hours, Ukraine's Deputy Prime Minister Mykhailo Fedorov tweeted directly at Musk, asking for Starlink terminals. Musk replied ten hours later: the service was active, and terminals were en route. The first shipment arrived on February 28—three thousand units flown via Poland.

By June 2023, the Ukrainian military was operating more than 42,000 Starlink terminals. They powered drones conducting reconnaissance and targeting, encrypted communications between forward units and command centres, and coordination of artillery strikes. Ukraine's army had become reliant on a service it did not own, operating satellites it could not access, under terms of service written in California.

The thing is, no legal framework anticipated this arrangement. The Outer Space Treaty of 1967 governs state activities in space and prohibits weapons of mass destruction in orbit, but says nothing about commercial constellations used for military purposes. The International Telecommunication Union allocates radio frequencies but does not regulate how satellite operators deploy or withdraw service. The result: Starlink operated in a governance vacuum. SpaceX made decisions that affected battlefield outcomes, and no multilateral body could compel it to act—or refrain from acting—in any particular way.

When the Switch Is in California

Dr. Juliana Suess, a space security fellow at the Royal United Services Institute in London, began tracking Starlink's role in Ukraine in early 2022. What struck her was not the technology—it was the decision-making authority. "You had a situation," she told me in February 2026, "where one company, accountable primarily to shareholders and U.S. export law, was making real-time choices about what a foreign military could and could not do. And those choices were being made in Hawthorne, not Kyiv or Brussels."

The geofencing incident Sharapov experienced was not isolated. According to documents reviewed by The Editorial, SpaceX implemented at least four separate geofence adjustments between March 2022 and November 2023, limiting Starlink service in areas SpaceX assessed as either Russian-occupied Crimea or contested zones where offensive operations might violate U.S. export restrictions on dual-use technology. Ukrainian officials protested each adjustment. SpaceX maintained that it was complying with U.S. law, which prohibits the export of certain technologies to support offensive military operations without explicit government authorization.

The issue came to a head in October 2022. Ukrainian forces planned a naval drone operation targeting Russian vessels in Sevastopol harbour in occupied Crimea. The drones required Starlink for navigation and remote piloting. Hours before launch, SpaceX disabled service in the area. Musk later stated publicly that he had refused to enable Starlink for offensive strikes deep into Russian-held territory, fearing escalation. Ukrainian officials were furious—they argued Crimea was sovereign Ukrainian territory under international law, and the operation was defensive in nature. But the decision had been made, and there was no appellate process.

The Leverage No One Anticipated

Here is what this means. Starlink's constellation is not merely infrastructure—it is geopolitical leverage. Countries that rely on it for military, government, or civilian communications cede a degree of sovereignty. They depend on a service that can be withdrawn, throttled, or geofenced by a private entity subject to U.S. law and corporate priorities. This dynamic extends far beyond Ukraine.

Taiwan's Ministry of Defense began evaluating Starlink in 2023 as a backup to undersea fibre-optic cables that China could sever in a conflict. But Taiwanese officials faced the same question Ukrainian commanders did: what happens if SpaceX decides, under pressure from Beijing or Washington, that supporting Taiwanese forces violates its terms of service? In a February 2025 closed-door briefing to Taiwan's Legislative Yuan, reviewed by The Editorial, defense officials acknowledged they had no contractual guarantee SpaceX would maintain service during a Chinese invasion. The ministry has since allocated $220 million to develop a sovereign low Earth orbit constellation, but deployment is not expected before 2029.

The leverage is not hypothetical. In August 2023, the Nigerian government requested Starlink service for rural areas. SpaceX agreed, but required approval from the Nigerian Communications Commission and the U.S. State Department. When Nigeria began deploying the terminals in regions where Boko Haram operated, SpaceX quietly implemented usage restrictions, limiting bandwidth during certain hours to prevent the terminals from being captured and repurposed by insurgents. Nigerian officials were not consulted before the restrictions were applied. They learned about them when field reports came back showing degraded service.

The Weapons Already in Orbit

The reliance on satellite constellations has not gone unnoticed by adversaries. Russia and China have both accelerated development of anti-satellite weapons—kinetic interceptors, co-orbital jammers, ground-based lasers, and cyberattack capabilities designed to disable or destroy satellites. The paradox is that destroying a mega-constellation is both easier and harder than destroying traditional satellites. Easier, because low Earth orbit satellites are numerous and more accessible than geostationary assets 35,000 kilometres up. Harder, because the sheer number creates redundancy—taking out one or two satellites accomplishes little when thousands remain.

But the debris problem is catastrophic. When China destroyed one of its own weather satellites in a 2007 test, the event created more than 3,400 trackable fragments that remain in orbit today, endangering every satellite that passes through that altitude band. A coordinated attack on Starlink—say, destroying fifty satellites—would generate tens of thousands of debris pieces, rendering large swathes of low Earth orbit unusable for decades. This is known as the Kessler Syndrome: a cascade of collisions that makes space inaccessible.

Dr. Brian Weeden, a former U.S. Air Force officer and director of program planning at the Secure World Foundation, has been studying counterspace weapons since 2009. "The introduction of mega-constellations fundamentally changed the ASAT calculus," he explained in a March 2026 interview. "If you're Russia or China, you now have to assume that Starlink or a similar system will be used against you in a conflict. That creates pressure to develop ways to deny or degrade that capability. But any kinetic ASAT attack risks debris that affects everyone, including your own satellites. So you get this unstable equilibrium where everyone is developing weapons, but no one wants to use them because the cost is planetary."

The Governance Vacuum

No treaty governs the use of commercial satellite constellations in warfare. The Outer Space Treaty prohibits placing nuclear weapons in orbit and claims of sovereignty over celestial bodies, but it does not address whether private companies can refuse service to belligerents, whether constellations constitute dual-use military infrastructure, or who adjudicates disputes when a company's terms of service conflict with international humanitarian law.

The United Nations Committee on the Peaceful Uses of Outer Space has been negotiating a framework for space resource governance since 2019, focused primarily on lunar mining and asteroid extraction. But the committee moves slowly—consensus requires agreement from eighty-four member states, including the United States, Russia, and China, who disagree on nearly every substantive question. As of April 2026, there has been no serious multilateral discussion of how to regulate mega-constellations used for military purposes.

The United States has proposed voluntary norms—such as the Artemis Accords, signed by thirty-six countries, which include commitments to transparent operations and debris mitigation. But the Accords are non-binding, and China and Russia have refused to sign, calling them a U.S.-led framework designed to entrench American commercial advantage. China has instead proposed its own International Lunar Research Station, a multilateral project that excludes the United States. Space governance is fragmenting along the same geopolitical lines as terrestrial governance.

What the Next War Will Look Like

Military planners in Washington, Beijing, and Moscow are now gaming scenarios where satellite constellations are the first target. A 2024 RAND Corporation study, commissioned by the U.S. Space Force, modelled a conflict over Taiwan in which China used cyberattacks and ground-based jammers to disable Starlink coverage over the Taiwan Strait in the first six hours. The study found that without satellite communications, U.S. and Taiwanese forces lost coordination advantages that could delay a Chinese amphibious assault by as much as seventy-two hours—time critical for reinforcements to arrive.

The study also found that SpaceX's ability to rapidly replace damaged satellites—the company launches dozens per month—created a kind of resilience traditional military satellites lack. But it also meant that any adversary would need to sustain attacks over weeks, not hours, to degrade the constellation meaningfully. That sustained campaign would generate debris, escalate the conflict into space, and likely trigger retaliatory strikes on the adversary's own satellites. The result: a war that begins on Earth quickly becomes a war that disables infrastructure both sides depend on for everything from weather forecasting to GPS navigation.

What We Still Don't Know

The fundamental question remains unresolved: who decides what happens in space when private infrastructure serves public—and military—purposes? Does SpaceX have the right to refuse service to a government at war? Does a government have the right to demand continuous service from a private company, even if that demand creates legal or financial risk for the company? And if a mega-constellation becomes a weapon by virtue of the role it plays in conflict, does that make it a legitimate military target under international law?

No international court has ruled on these questions. No treaty addresses them. And the precedent being set right now—in Ukraine, in Taiwan's planning rooms, in Nigeria's rural deployments—is that the answers will be determined ad hoc, by individual companies, under domestic law, with no multilateral oversight.

Lieutenant Colonel Sharapov still uses Starlink. His unit received newer terminals in January 2026, with updated encryption and faster handoff between satellites. The service works most of the time. But he has stopped assuming it will work when he needs it most. "We plan for it to disappear," he said in a brief encrypted message exchange in March. "We train for the moment the screen goes black. Because we know now that the decision to keep it on is not ours to make."

The war in Ukraine demonstrated that space infrastructure is now battlefield infrastructure. The governance frameworks designed for an era when only governments could reach orbit have not caught up. And the companies building the constellations are moving faster than the diplomats trying to regulate them. What happens when the next war begins, and the satellite network everyone depends on is controlled by a handful of corporations accountable to no multilateral body, is a question we are answering in real time—one geofence, one terms-of-service update, one outage at a time.