Until recently, governments controlled the means to wage war. No longer. Private companies own and manage critical military infrastructure like satellites, cloud computing, and artificial intelligence. Call it the switch power. For policymakers, the shift in control deserves attention.
The clearest early signal of switch power came in Ukraine. Ukrainian forces were preparing a naval drone strike against Russia’s Black Sea Fleet off occupied Crimea. The operation depended on Starlink. Elon Musk declined to activate it, arguing the strike risked excessive escalation. No government could force immediate activation; no alliance mechanism stepped in; Kyiv had no recourse beyond persuasion.
It’s a dangerous dependency. Even though most of the key switch companies are American, they often balk at following orders and find themselves in conflict with the US government — or held hostage by hostile governments. For Europe, reliance on US companies sits awkwardly alongside ambitions for strategic autonomy.
Competition to the Starlink monopoly is emerging. Amazon is building what it now calls Amazon Leo, formerly Project Kuiper; Europe is investing in its own Starlink rival IRIS². France recently hosted an International Space Summit to build Europe’s sovereign space capabilities. The US is attempting to carve out a secure military use of Starlink called Starshield.
But dependence has grown faster than alternatives can be built. In the meantime, governments would do well to treat this as a standing condition rather than a one-off failure to be fixed and forgotten.
Governments have relied on defense contractors for centuries without losing strategic control over the relationship. Switch power requires four conditions:
- The infrastructure is operationally central, not merely helpful.
- No comparable substitute exists.
- No substitute can be deployable in time.
- The provider retains real discretion over access, pricing, coverage, or use conditions, rather than simply executing fixed instructions.
Starlink met these conditions during the Crimea strike: it linked navigation, command, targeting, and execution into a single chain, and no alternative could have been fielded on the relevant timeline.
In 2026, Ukrainian officials said SpaceX disabled the Starlink access that Russian drones depended on. Washington received a lesson during operations against Iran when SpaceX reportedly raised prices for military satellite services as demand grew, and the Pentagon paid because replacing that capability mid-campaign was unrealistic.
A private company not only can change prices or contract terms mid-conflict. It can pick sides, or even carry out policies its country of origin opposes. Hostile governments can apply pressure by closing market access or imposing regulations that hurt businesses. Sometimes private companies refuse orders. The Pentagon ended up in litigation against Anthropic after the AI company refused contract terms permitting mass surveillance and use of autonomous weapons — a dispute that left a procurement decision hostage to a courtroom timeline rather than an operational one.
Taiwan is where the stakes are clearest. Beijing does not need to attack American satellites directly to blunt US military effectiveness around the island; it can lean instead on the commercial exposure of firms doing business in China. Chinese authorities froze expansion plans for Tesla’s Shanghai factory, citing data security concerns tied to the rollout of Starlink over Ukraine.
Redundancy needs to be operational: a backup provider listed in a procurement annex is not the same as one that can actually assume a mission mid-conflict. Contracts should treat political refusal as seriously as technical failure, spelling out crisis-activation terms, territorial restrictions, and data portability with the same rigor applied to cost and delivery schedules. Militaries should rehearse not only the loss of a platform or a network, but the loss of the company that operates it.
Few defense ministries currently have such a plan on file. Governments will still decide when they want to fight. Increasingly, though, their freedom to act is conditioned by who holds a switch. That is a question most governments have not yet asked — or answered.
Nikola Lunić is a Serbian analyst and Navy Captain (Ret.). He previously served as Serbia’s Defence Attaché in London and Oslo and as Executive Director of the Council for Strategic Policy. He is a Strategic Affairs Consultant and a regular guest lecturer at the Faculty of Law, University of Osijek. His analysis has appeared in CEPA, New Eastern Europe, and Kyiv Post, among other outlets.
Bandwidth is CEPA’s online journal dedicated to advancing transatlantic cooperation on tech policy. All opinions expressed on Bandwidth are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
Tech 2030
A Roadmap for Europe-US Tech Cooperation
