Satellite Internet is not a race anymore but a rout where two losers are still figuring out what game is being played. First off, here’s a number that makes any discussion pointless – Starlink currently has about 10,700 operational satellites in the sky out of an intended constellation of up to 42,000 satellites. Amazon Leo (rebranding of the Project Kuiper) has reached 375 satellites after 14 orbital launches. And OneWeb owned by Eutelsat currently has 648 satellites. Another way to put it is this – Starlink already has almost fifteen times more satellites in space than both nearest competitors from the Low Earth Orbit.
Also read: Mark Zuckerberg admitted that Meta’s bet on AI Agents isn’t working, here’s why
According to the FCC, the Starlink constellation has permission to deploy a maximum of 15,000 satellites by the end of January 2026, with an aim of deploying up to 42,000 satellites going forward. New satellites from SpaceX are launched by the Falcon 9 spacecraft at an interval of a few days, which Amazon and Eutelsat cannot match due to the fact that both lack control over their own rocket fleets like SpaceX.
Amazon Leo appears the most plausible contender in theory; it is the worst-case study on how “in theory” does not work. It has reached 375 satellites and is now the third biggest constellation of satellites orbiting the Earth, having 100 planned launches on five rocket families, namely Atlas V, Ariane 6, Vulcan Centaur, New Glenn, and, importantly, SpaceX’s very own Falcon 9. The act of Amazon spreading out on a competitor’s rocket while competing with that competitor’s internet services is quintessentially Amazon.
Also read: WhatsApp username row in India isn’t about privacy vs security, but authentication
The issue is timing. Amazon’s own FCC license stipulates 50% of the first generation of constellations, which is around 1,618 satellites, to be placed in orbit by July 30, 2026. It will miss the deadline and had to ask for relief from the deadline, which the FCC gave it in June 2026, imposing penalties in the form of low spectral priority for all future satellites.
OneWeb completed the construction of its first generation of satellites in 2023, becoming the first company to deploy an entire LEO broadband constellation and retaining its title as the second largest in operation today. However, it was not designed to compete against Starlink for your home WiFi. OneWeb is oriented towards airlines, shipping companies, the military, and governments, and orbits at a high and near-polar level that is geared toward coverage, not speed. Eutelsat will purchase 440 satellites of the next generation from Airbus to replace its aging fleet beginning in late 2026, but it’s a business story dressed up in satellite internet garb.
If you’re asking who wins the race to blanket the planet in broadband, Starlink already has. Nothing on Amazon’s or Eutelsat’s launch manifest changes that arithmetic before the decade is out. If you’re asking who wins on interesting business strategy, it’s Amazon, if only because betting on five separate rocket families, including a competitor’s, is either genius risk management or a confession that Blue Origin’s New Glenn wasn’t ready to carry the load alone.
And if you’re asking who wins the fight for the niche that actually pays, that’s OneWeb, quietly profitable in a lane Starlink and Amazon are still figuring out how to enter. Three different races, three different scoreboards. Only one of them is a rout.
Also read: Anthropic’s controversial data retention for Fable 5 and Mythos 5