On March 23, 2026, a Soyuz-2 rocket lifted off from the Plesetsk Cosmodrome carrying 16 operational low-Earth-orbit satellites for Bureau 1440, the Russian company building a constellation called Rassvet ("Dawn"). According to an April assessment by the Ukrainian defense outlet Militarnyi, the prototype and first-batch satellites already pass directly over Ukraine two to three times a day, opening communication windows of roughly 15–20 minutes each. Bureau 1440 plans 156 satellites in orbit during 2026 and 292 by 2027 — the threshold for full commercial service. For the first time, Russia is on a credible path to a domestic Starlink.
The strategic logic is uncomfortable but clear. Ukraine's battlefield connectivity has depended on a single American company, and Russia has spent three years without an equivalent. Rassvet changes that. Militarnyi warns that the terminals — roughly 60 cm, under 15 kg, rated for field temperatures — are likely to be adopted quickly by Russian forces and fitted to Geran and Garpia strike drones and cruise munitions. Crucially, unlike Starlink, a Russian-operated network carries no geofencing: SpaceX has restricted service near contested areas and, in 2023, declined to enable Starlink over Crimea for a Ukrainian operation. A sovereign Russian constellation answers to the Kremlin, not to a corporate terms-of-service page.
The dependency cuts both ways
It would be easy to read Rassvet purely as a Russian capability story. The sharper lesson is about Ukraine's own unresolved exposure. Kyiv operates more than 50,000 Starlink terminals across military, medical, and civilian networks, as the Kyiv Independent reported when Germany began financing access to a European alternative. That single-vendor reliance has been a recurring vulnerability — subject to one founder's decisions, one government's aid posture, and one company's commercial priorities. Russia building its own resilient network does not create Ukraine's dependency; it raises the cost of leaving it unaddressed.
The steelman for heavy state involvement here is real. Strategic connectivity is plausibly a public good that markets under-supply: no commercial operator will guarantee wartime coverage at a loss, accept sovereign control over kill-switch decisions, or accredit its security architecture to a defense ministry's standards. That is precisely why the EU is funding IRIS², its sovereign multi-orbit constellation of about 290 LEO and MEO satellites, and GOVSATCOM, a pooling arrangement that lets member states share secure satellite capacity. Both are run by public institutions for exactly the assurance a private vendor cannot offer.
But Europe's timeline is the problem
Here the pro-innovation critique bites. IRIS² is, by the European Commission's own page, "expected to become operational by 2030." GOVSATCOM remains a capacity-sharing framework without a published deployment date, and Ukraine's access to it still requires a formal international agreement and EU security accreditation of the GOVSATCOM hub. Russia, meanwhile, is targeting commercial Rassvet service in 2027. A procurement-led European answer that arrives three years after the threat matures is not proportionate to the operational reality.
The near-term substitute, Eutelsat OneWeb, illustrates the gap. Germany is financing Ukraine's use of Eutelsat, with the operator aiming to grow from under 1,000 terminals to 5,000–10,000 "relatively fast." That is complement, not replacement: against 50,000-plus Starlink units, OneWeb's capacity ceiling and far higher per-terminal cost make it a hedge rather than a swap. Europe is right to build sovereign infrastructure; it is moving too slowly and too monolithically to field it in the window that matters.
What proportionate policy looks like
The instinct in Kyiv and Brussels will be to nationalize the answer — one flagship EU constellation, procured centrally, online by 2030. That risks repeating the single-point-of-failure mistake at a continental scale, with a slower clock. A more resilient and more innovative posture treats diversity itself as the security property:
- Multi-vendor by design. Fund parallel access to Starlink, Eutelsat OneWeb, and IRIS² as it comes online, so no single provider — or its owner's politics — is a chokepoint. Redundancy across operators is what survives both a kill-switch and a jammer.
- Buy now, build in parallel. Treat 2026–2029 as a bridge to be filled with whatever commercial capacity exists, rather than waiting for the sovereign system. The cost of interim Eutelsat terminals is trivial next to the cost of a coverage gap.
- Accelerate the accreditation path. The binding constraint on GOVSATCOM for Ukraine is institutional — international agreements and security sign-off — not orbital. That is a problem European policymakers can fix on a political timeline, not an engineering one.
- Keep terminals open and interoperable. Avoid locking Ukrainian forces into one proprietary ground segment. Interoperable, multi-network user terminals are the hardware expression of vendor diversity.
Rassvet is not yet a finished system — Bureau 1440 lost one of the first-batch satellites within months of launch, and 16 spacecraft do not make a constellation. But the direction is unmistakable, and the asymmetry it exposes is genuine: Russia is industrializing a geofence-free military network on a 2027 clock, while Ukraine's allies debate a sovereign answer pencilled in for 2030. The right response is not a single bigger public project. It is faster procurement, deliberate redundancy, and a refusal to let any one satellite operator — American, European, or otherwise — become the war's single point of failure.