Starlink’s rapid growth is transforming Elon Musk's SpaceX into an internet and national security powerhouse even as rivals race to challenge its lead
Alan Dyer/VWPics/Universal Images Group via Getty Images
A version of this article originally appeared in Quartz’s members-only Weekend Brief newsletter. Quartz members get access to exclusive newsletters and more. Sign up here.
Starlink just crossed nine million customers, less than two months after hitting eight million. SpaceX launched the service in 2019 as a bet that low-Earth-orbit satellites could deliver broadband fast enough to matter in places cable had never reached.
Join 500,000+ readers who start their day with Quartz.
By subscribing, you agree to our Terms of Service and Privacy Policy.
The internet service now operates in 155 countries, territories, and markets, with more than 9,000 satellites in orbit. It adds about 21,000 new users every day, the kind of upward curve that helps explain why SpaceX is reportedly aiming to go public this summer at a valuation near $1.75 trillion.
Elon Musk has called it "by far" SpaceX's largest revenue driver, and the IPO paperwork reviewed by Reuters backs him up. Starlink doubled its operating income last year to $4.42 billion, more than enough to cover losses in SpaceX's rocket and space division.
That growth has turned Starlink into the financial engine for everything else SpaceX wants to build. It has also turned it into a target.
Cable companies stopped running fiber when the math stopped working, which left a lot of people behind, including ranchers in Montana, container ships off Singapore, regional jets at 35,000 feet, and the remote workers who scattered during the pandemic and stayed scattered.
Satellites were not obviously cheaper until SpaceX figured out how to land its rockets and reuse them. Once it did, beaming internet down from low orbit started to pencil out, and Starlink got to all of those places at once. And the world seems hungry for more. Fast, consistent internet no matter where you are has gone from a nice thing to have to something closer to a necessity.
Washington has helped Starlink along. In January, the FCC granted SpaceX a partial approval to operate 7,500 more second-generation satellites and fly them at lower orbits, which means lower latency, and a separate waiver lets Starlink exceed federal power limits to push gigabit download speeds, roughly five times what most subscribers get today.
The Pentagon has helped even more. SpaceX has joined a nine-company group building software for the Trump administration's Golden Dome missile shield, a program estimated at $175 billion, with Starlink positioned to be a key piece of the satellite communications layer.
Add in a $178.5 million Space Force contract from April and the billions in government contracts the company got in 2025, and Starlink looks less like a consumer internet service and more like connective infrastructure for U.S. national security in space.
Growth like that, plus the Pentagon money behind it, has the sharks circling. Amazon $AMZN is the biggest one.
The company has been working on its own low-Earth-orbit constellation, rebranded from Project Kuiper to Amazon Leo, for several years, but it currently has only around 200 satellites in orbit against Starlink's 9,000.
The satellite gap will be hard to close, but Amazon's $11.57 billion deal to acquire Globalstar gets it something arguably more valuable: spectrum, the radio frequencies that make satellite-to-phone service work, and which the FCC just denied SpaceX more of. The deal also brings a global network of ground stations and the contract behind Apple $AAPL's iPhone SOS service. Amazon plans to deploy a next-generation system in 2028 and has already lined up major commitments from airlines, telecoms, and NASA.
Jeff Bezos is coming for Starlink from a second direction too. His rocket company Blue Origin is planning its own internet constellation, called TerraWave, aiming to launch at least 5,400 satellites by the end of 2027 with a focus on large business customers.
China presents the longer shadow. A newly formed Chinese institute filed paperwork in late December for two constellations totaling roughly 193,000 satellites, dwarfing the 49,000 that SpaceX has proposed. Beijing has been slow to actually launch anything, but the FCC has framed the buildout as a national security threat. The FCC's top satellite regulator, Jay Schwarz, called it "space race 2.0" in a November speech.
There is a case that the competition is overdue. Starlink today routes through a single CEO who has changed his mind on service in active war zones in Ukraine, and whose network Iran has figured out how to jam inside its own borders. The Pentagon is buying connective infrastructure whose on-off switch sits with one person, and that person is increasingly distracted by xAI, humanoid robots, and a growing pile of side bets. Amazon, Blue Origin, and the Chinese consortium (for Beijing at least) represent something no defense apparatus currently has, which is an alternative.
But too much competition would ruin the plan. The IPO will test whether investors believe Starlink's lead is durable enough to fund everything Musk has stacked on top of it, including a pay package that could end up worth even more than the eye-popping trillion-dollar deal Tesla $TSLA shareholders just approved. The SpaceX board has tied his full payout to the company reaching a $7.5 trillion valuation and putting a million people on Mars. Musk's favorite planet now has more reasons than ever to stay his favorite.
