A cluster of San Francisco startups is trying to do to satellite constellations what Amazon Web Services did to server racks: turn them into something customers can use without owning the whole technical stack. The pitch is that a company, government, research group or media organization should be able to task orbiting hardware directly, without negotiating with Starlink, Maxar, Planet Labs or a national space agency every time it needs data or connectivity.
Whether that pitch can actually dent Starlink’s grip on low Earth orbit is the question worth examining. The evidence so far suggests not, at least not on the terms the headlines imply. What Basalt Space and its peers are really selling is something narrower and more interesting: sovereignty, not scale.
Apartments as factories
Basalt Space operates with the aesthetic of a hacker house colliding with a satellite factory. WIRED recently reported that Max Bhatti and four other Basalt engineers worked 22-hour days in March to assemble the startup’s first satellite before a launch deadline, using adjacent San Francisco apartments as home, office and workshop.
The aesthetic is deliberate. The company’s argument is that one of the aerospace industry’s most important assumptions can be broken: that end users cannot directly task a constellation of their own. Basalt’s public pitch is built around autonomous, tailored satellite constellations for private enterprise and government customers. Y Combinator describes the company as offering “Constellations as a Service,” enabling users to fly and stream data from their own tailored satellite constellation from a phone or laptop.
That matters because the ownership model is the point. The customer is not merely buying an image, a signal or a bandwidth slice from somebody else’s network. They are buying control over orbital capacity that is supposed to be theirs to task, reprioritize and secure.
That pitch is aimed squarely at governments and corporations that watched Starlink’s role in Ukraine and concluded that depending on a single commercial gatekeeper for orbital connectivity can become a strategic liability.
Why now
Three forces converged to make small constellations economically plausible. SpaceX’s rideshare program lowered launch costs for small payloads. Commercial electronics and space-qualified components became more capable and available. And U.S. regulators have been moving to speed parts of the commercial space approval process.
The policy backdrop matters. In August 2025, the White House issued an executive order directing federal agencies to streamline commercial launch and reentry approvals and expedite or eliminate some environmental reviews where legally possible. The FCC separately adopted rules to expedite initial processing of satellite and earth-station applications, including measures aimed at reducing burdens for the ground-station-as-a-service market.
The historical contrast is sharp. A satellite carrying the infrared sensors and high-speed downlink needed to monitor wildfires would once have been a large, expensive government or defense-prime project. Today, smaller companies can build far cheaper versions, launch them on rideshare missions, and sell the resulting data or control layer directly to customers.
Astranis, another Bay Area entrant, is making a related sovereignty pitch from a different orbit. Rather than building low-Earth-orbit swarms, it sells small geostationary broadband satellites to customers, including governments and telecom operators, that want dedicated national or regional connectivity. The demand is framed bluntly: countries and large institutions want private, secure networks and greater confidence about who controls their data.
Starlink’s shadow, and the gap it leaves
There are now well over 10,000 operating satellites in Earth orbit, with broader counts that include inactive satellites running closer to 15,000. The Satellite Industry Association reported that 11,539 satellites were operating in Earth orbit at the end of 2024, up from 3,371 in 2020. Popular-science tallies that include inactive satellites put the total number orbiting Earth around 15,000.
That distinction matters because Starlink dominates the active low-Earth-orbit broadband layer. SpaceX has launched thousands of Starlink satellites and has become the default reference point for what scale looks like in commercial LEO. Quilty Space has forecast that roughly 20,000 satellites will launch by the end of the decade under a more conservative, funded-projects-based model, with low-Earth-orbit broadband megaconstellations making up the bulk of Western demand.
That concentration is precisely the condition the challengers are pitching against, and also the reason the “end the monopoly” framing oversells what they can plausibly do. Starlink’s manufacturing cadence, launch access, customer base and operational experience are not being matched by apartment-born startups. The realistic ambition is not to displace it. It is to carve out a parallel market for customers who want exclusive control of their own orbital hardware.
The bet Basalt and Astranis are making is that a meaningful slice of customers, including sovereign governments, defense agencies, regulated industries, insurers, logistics firms and critical-infrastructure operators, will pay a premium to opt out of dependence on a shared commercial network. Government demand is rising in parallel. The U.S. Space Force’s Proliferated Warfighter Space Architecture and Europe’s IRIS² program both signal that the buyers most willing to pay for sovereignty are already convinced that orbital access is strategic infrastructure.
The military shadow
The democratization narrative has a darker counterweight. A growing share of commercial constellation revenue comes from defense and intelligence customers. The Wall Street Journal recently reported that for satellite startups, war pays better than climate monitoring, as companies that once emphasized agriculture, sustainability or environmental monitoring increasingly pivot toward defense contracts.
Bloomberg has documented a related problem on the imagery side: commercial satellite providers can restrict access around active conflict zones, especially when governments ask them to withhold imagery. Planet Labs’ decision to restrict Middle East imagery during the Iran conflict raised questions about commercial independence and global accountability in a market often described as open or democratized.
Civil liberties researchers have also flagged the convergence between commercial space and the surveillance state, with Tech Policy Press documenting how the new space economy is increasingly built around intelligence-community contracts and defense demand. The lease-your-own-constellation pitch sounds liberating until the question becomes: who is leasing, and to do what?
Space Daily has previously covered how orbital startups have rapidly pivoted toward defense applications, with valuations following Pentagon interest rather than pure commercial traction. That pattern reinforces the narrower reading of what these companies are: not Starlink killers, but bespoke orbital contractors for buyers who consider Starlink itself a geopolitical risk.
What could break the model
The risks are concrete. Demand has to actually show up. Many of the companies betting on a flat-rate constellation market may discover that customers prefer to buy data from incumbents rather than operate hardware they do not understand, even if the operating layer is wrapped in software.
Then there is the orbital environment itself. Even conservative satellite-growth forecasts represent a meaningful increase over today’s population, with collision risk, debris, spectrum congestion and astronomical interference scaling alongside. Nature and other scientific outlets have warned that satellite swarms are already changing astronomical observations, and proposed megaconstellations could make that problem far worse.
Regulators have so far leaned toward permissiveness and speed, especially in the United States. Whether that holds after the first major in-orbit collision involving a commercial constellation, or after a serious spectrum-conflict dispute between satellite operators, is an open question.
The deeper shift
Strip away the Bay Area packaging and what is happening looks less like the end of Starlink’s dominance and more like the unbundling around it. A capability that once lived only inside governments and a handful of defense primes is being repriced and offered to anyone with a purchase order. But the mass-market broadband layer is already taken.
The question worth watching is not whether Basalt or Astranis displace Starlink. They almost certainly will not. It is whether their alternative model, orbital infrastructure controlled by its users rather than rented from a gatekeeper, establishes itself firmly enough that sovereign customers stop treating Starlink as the default.
If it does, the geopolitics of space fragments along familiar lines: an American commercial backbone, plus a constellation of national, corporate and mission-specific alternatives that exist precisely because depending on that backbone has become politically uncomfortable. If it does not, consolidation around Starlink and a few state-backed competitors will harden into something closer to permanent.
Photo by Chris Lyo on Pexels
