+
Elon Musk just put a breathtaking new number on Starlink’s long-term ambitions: more than 100,000 satellites.
That would be far beyond the constellation flying today—and far beyond what regulators have currently authorized.
But Musk’s bigger claim may be the one attached to those spacecraft. He says future Starlink satellites could deliver more than 50 times the throughput of the network’s V2 design.
Musk laid out the scale directly Sunday, saying the total includes satellites built for direct-to-cell service.
Yes, I expect the Starlink total constellation size, inclusive of direct-to-cell, to exceed 100k.
Future versions of our satellites may exceed 50X the throughput of a V2, making the total system bandwidth even higher.
— Elon Musk (@elonmusk) August 9, 2026
The post is a statement of direction, not a deployment schedule. SpaceX would need more launch capacity, more spectrum and significant regulatory approval before a constellation on that scale could become real.
Still, the number shows how differently Musk sees Starlink’s future. This is no longer framed as a satellite-internet service that merely fills dead zones.
It is becoming a global communications layer for homes, aircraft, ships, phones—and eventually a huge population of machines talking to other machines.
The Federal Communications Commission gave SpaceX a major piece of that runway in January after reviewing the company’s proposed Gen2 upgrades and orbital changes. The agency authorized an additional 7,500 Gen2 Starlink satellites, bringing the approved Gen2 total to 15,000 worldwide.
That order also opened more operational flexibility across fixed and mobile satellite services. It allowed SpaceX to use additional Ku-, Ka-, V-, E- and W-band frequencies, added orbital shells between roughly 340 and 485 kilometers, and expanded the path for mobile and direct-to-cell coverage.
The FCC’s action does not authorize a 100,000-satellite network. SpaceX’s broader Gen2 application sought 29,988 satellites, and the agency chose to move incrementally.
Musk’s new target therefore sits well beyond both the current grant and that earlier application.
The agency said the authorized expansion is meant to improve high-speed, low-latency broadband while supporting mobile and supplemental coverage from space. It also tied lower-orbit operations to conditions designed to protect other spacecraft and NASA missions.
SpaceX’s engineering answer is Starlink V3.
An official Starlink update describes V3 as a next-generation satellite built around higher capacity, greater data density and more power generation. The company says those gains, combined with Starship’s much larger payload capacity, are what can let the constellation scale while serving millions more customers.
That last part is crucial. A bigger satellite is useful only if SpaceX can put enough of them into orbit at a workable cost. Starship is meant to carry far more Starlink hardware per flight than Falcon 9, turning deployment from batches of a few dozen into something much larger.
And raw satellite count is only half the story. If later spacecraft really deliver multiples of V2 throughput, the network could add capacity much faster than the headline number alone suggests.
Musk also pointed to the demand he expects to consume it.
AI agentic Internet traffic will obviously VASTLY exceed human usage. Not a close call at all.
Cloudflare’s forecast is accurate. https://t.co/VztgrinN5k pic.twitter.com/Wo4FiRKjPU
— Elon Musk (@elonmusk) August 9, 2026
That forecast changes the shape of the internet. Human browsing, streaming and video calls are heavy, but AI agents can work continuously—querying services, moving data, generating media and coordinating with other systems around the clock.
A network designed for that world would need enormous capacity and broad reach. Starlink has both a space-based backbone and a direct-to-cell path, while SpaceX now has an explicit reason to push bandwidth far beyond ordinary consumer demand.
There are hard questions ahead. A constellation of 100,000-plus satellites would intensify debates over orbital congestion, astronomy, debris mitigation and international spectrum coordination.
SpaceX will have to prove that its space-traffic surveillance, collision-avoidance and deorbit systems can scale with the network.
But the ambition is unmistakable.
Starlink began by bringing broadband to places cables and towers could not reach. Musk now sees something much larger: a communications system with more than 100,000 spacecraft, orders of magnitude more capacity and enough bandwidth for an internet increasingly populated by machines.
That is a long road from today’s authorization. It is also the clearest picture yet of where Musk wants Starlink to go.
Join the conversation!
Please share your thoughts about this article below. We value your opinions, and would love to see you add to the discussion!