For two years the standard answer to “is Starlink going to become a real phone carrier” has been a polite no. The satellites were impressive, the physics were unforgiving, and the service you could actually buy was a safety net for the places where coverage runs out. Texting from a canyon, not streaming from one.
On October 8, 2026 SpaceX changed the shape of that answer. The company announced an agreement to buy a nationwide low band spectrum portfolio, and then said out loud what it had only hinted at before: that it intends to build “the world’s most advanced mobile network” and become “a major mobile carrier in the US.”
The amount of spectrum involved is 14MHz. That number is small enough to be worth sitting with for a moment, because it explains both why this announcement matters and why the big three carriers are not finished yet.
What SpaceX actually bought
The seller is Grain Management, an investment firm that acquired the 800MHz portfolio from T-Mobile in a transaction that closed in August 2026, paying roughly $2.9 billion in cash along with its own 600MHz holdings. Barely two months later Grain has agreed to pass those licenses to SpaceX.
The portfolio is nationwide and covers up to 14MHz of paired spectrum in the 800MHz band. SpaceX did not disclose what it is paying, and the deal cannot close until the FCC signs off. Elon Musk described the acquisition as “the last critical piece of the spectrum puzzle” for phone coverage in America, which is a characteristically confident framing of a portfolio that is modest by carrier standards.
To understand why SpaceX wanted it anyway, you have to look at what the company already had.
| Holding | Amount | What it is good at | Status |
|---|---|---|---|
| 2GHz mid band (from EchoStar) | About 65MHz | Capacity and throughput. Carries real bandwidth. | Acquired in a roughly $17 billion deal, with full transition running into 2027 |
| 800MHz low band (from Grain) | Up to 14MHz paired | Reach and building penetration. Gets a signal indoors. | Announced October 8, 2026, pending FCC approval |
| Gen2 satellite authorization | 15,000 satellites total | Coverage where no tower exists at all. | Granted by the FCC, with half required to launch by December 1, 2028 |
| Terrestrial network | None yet | Dense urban capacity and indoor coverage. | Described only as a plan |
Read down that table and the strategy is legible. SpaceX has bandwidth, it now has reach, it has permission to fly a very large constellation, and it has essentially no ground infrastructure. The first three are purchases and filings. The fourth is years of construction.
Why 14MHz of 800MHz is worth buying
Spectrum is not interchangeable, and the difference between SpaceX’s two bands is not a detail. It is the whole reason this deal happened.
A radio wave at 2GHz carries a lot of data and stops at the first serious obstacle. Walls, windows with modern coatings, elevator shafts and office cores all eat it. A radio wave at 800MHz carries less data and travels much further, bending around terrain and pushing through building materials that would stop the higher band outright. Every terrestrial carrier in the United States runs both kinds for exactly this reason: low band for the blanket, mid band for the capacity.
Starlink Mobile has been missing the blanket. Its existing 2GHz mid band position is genuinely useful for throughput, but a satellite signal arriving from orbit at 2GHz has already spent its budget getting through the atmosphere. Asking it to also get through a brick wall is asking too much.
There is a second, less discussed reason low band matters to a satellite operator. Lower frequencies are more forgiving of the enormous distances and awkward angles involved in talking to a phone from low Earth orbit, which is why the first generation of direct to cell services leaned on repurposed low band carrier spectrum in the first place. Owning its own nationwide low band licenses means SpaceX no longer has to borrow anyone’s.
The satellite half of the plan
The spectrum news arrived alongside renewed attention on SpaceX’s Gen2 constellation authorization, and this is where a lot of October coverage got muddled, so it is worth separating.
The FCC granted SpaceX authority on January 9, 2026 to deploy an additional 7,500 Gen2 Starlink satellites, bringing its authorized Gen2 total to 15,000. SpaceX had asked for roughly 30,000 and the Commission deferred the remaining 14,988. The grant carries a deadline: half the newly authorized satellites must launch by December 1, 2028 or the authorization is at risk. Several articles published this week described that approval as brand new. It is not, and the 15,000 figure is a cumulative ceiling rather than a fresh batch.
What is genuinely new is the capability SpaceX is claiming for the next generation hardware. The company says the new satellites can deliver more than 100 times the bandwidth of the current direct to cell generation, a figure that comes from pairing much larger satellites with the EchoStar mid band spectrum. Those satellites are not expected to begin deploying until 2027.
How 14MHz compares to a real carrier
Here is the part that keeps this story honest. Independent analyses of carrier spectrum depth, measured as a population weighted average across the country, put low band holdings for the big three in the neighborhood of 47MHz to 56MHz each. Those are estimates from third party analysis rather than official filings, and they move as deals close, but the order of magnitude is not in dispute.
That gap is not fatal, because SpaceX is not trying to replicate a terrestrial network. Its argument is that a constellation plus a thinner ground layer can cover the country more cheaply than tens of thousands of conventional macro towers. Gwynne Shotwell has previously described the approach as using a large number of lower cost base stations paired with Starlink dishes instead of traditional cell sites, which is a genuinely different cost structure rather than a cheaper version of the same one.
Whether it works is an open engineering question, and the people who do this for a living are not sold yet.
The towers are the missing piece
Tim Farrar of TMF Associates put the problem plainly to CNBC: “This adds fuel to the fire in the battle between SpaceX and the mobile operators, but it’s still a very limited amount of spectrum, and to get reliable building penetration in urban areas, SpaceX would have to deploy towers on the ground.”
SpaceX does not dispute the second half of that. Its own announcement describes deploying the new spectrum through “an advanced terrestrial deployment,” which means radios and antennas on towers, rooftops and other structures. The company has not said how many sites, where, on what schedule, or at what cost.
This is the gap between a spectrum portfolio and a carrier. Licenses can be bought in an afternoon. Site acquisition, zoning, backhaul, power and construction take years, and they are the part of the business that the incumbents have spent three decades and hundreds of billions of dollars building. The 800MHz licenses themselves arrived with construction obligations when the FCC approved T-Mobile’s sale to Grain, and whether those conditions carry over unchanged to SpaceX is one of the things the Commission will decide.
What this means for your phone
Nothing, for now, and that is worth being clear about.
There is no Starlink Mobile retail plan to sign up for, no pricing, and no launch date. The FCC has to approve the Grain transaction. The EchoStar mid band spectrum has its own multi year transition timeline. The satellites that would make a real service possible start flying in 2027. A terrestrial network that barely exists on paper has to get built.
What has changed is the credibility of the plan. A year ago the case against Starlink becoming a carrier was that it did not own the right spectrum and could not operate without leasing from an incumbent. SpaceX has now addressed both: it holds mid band for capacity, it is acquiring low band for reach, and an FCC waiver lets it sell wireless service without leasing spectrum from a ground based carrier. The remaining objection is execution, which is a very different kind of objection.
If you want the fuller picture of how the company got here, we walked through the strategy behind SpaceX’s push to become your wireless carrier earlier this year, and most of what was speculative then has since been bought and filed.
Frequently asked questions
Can I get Starlink Mobile as my phone plan right now?
No. There is no retail Starlink Mobile service in the United States. Satellite connectivity on consumer phones today comes through carrier partnerships, principally T-Mobile’s service in the US, or through handset makers like Apple. SpaceX has announced ambitions and acquisitions, not a product.
Will my current phone work with Starlink Mobile?
Partly, and this is an underrated catch. Existing direct to cell service works on ordinary phones because it uses bands your phone already supports. The 800MHz licenses fall in territory most modern phones can handle, but the EchoStar 2GHz spectrum needs chipset and band support that many phones in pockets today do not have. A full Starlink Mobile service running on all of its spectrum will likely want newer hardware.
How is this different from what T-Mobile already offers?
Today’s T-Mobile satellite service is SpaceX capacity sold through a carrier, using the carrier’s spectrum, as a supplement to its terrestrial network. What SpaceX is describing is its own network, on its own spectrum, sold directly, with satellites as a primary layer rather than a backstop. One is a partnership. The other is competition.
Is 14MHz really enough to run a carrier?
Not on its own, and SpaceX has not claimed otherwise. It is enough to provide a coverage layer that reaches indoors, which the company did not previously have. The capacity has to come from the 2GHz mid band and the satellites, and the dense urban performance has to come from ground infrastructure that does not exist yet.
When could a real service launch?
No date has been announced. The constraints that are public point to 2027 at the earliest for next generation satellites, with the EchoStar spectrum transition running into that year as well. A nationwide service competitive with the big three on in building coverage depends on a terrestrial buildout that has no published timeline at all.
Does satellite service cost extra?
It depends entirely on the provider, and pricing in this category is unsettled. Apple, for instance, has now extended free satellite features on iPhone for four years running without ever naming a price for what happens next. Carriers have taken varied approaches, bundling satellite texting into some plans and charging for it on others.
The bottom line
SpaceX bought a small amount of exactly the right spectrum, and then made a very large claim about it. Both halves of that sentence are accurate, and the gap between them is the story.
The 800MHz portfolio solves a real problem. Starlink Mobile could reach the open sky and not much else, and low band is the only thing that gets a signal through a wall. Combined with the mid band from EchoStar, the Gen2 authorization and a regulatory waiver that lets the company sell service on its own terms, SpaceX has assembled every piece a carrier needs except the expensive one.
That last piece is the ground. Until SpaceX starts naming sites and pouring concrete, “major mobile carrier in the US” describes an intention rather than a network. The incumbents have a few years of breathing room, and they should probably spend it looking up.
Sources and further reading
- TechRadar: Starlink confirms plans to become the world’s most advanced mobile network and a major US carrier
- Via Satellite: SpaceX moves to acquire nationwide low band spectrum for Starlink Mobile
- Converge Digest: SpaceX agrees to buy nationwide 800MHz spectrum
- EchoStar: Spectrum sale and commercial agreement with SpaceX
- FCC: Order granting additional Gen2 Starlink satellites
- RCR Wireless: SpaceX wins FCC approval for 7,500 additional Starlink satellites
- The Register: SpaceX bulks up Starlink Direct to Cell with $17B EchoStar spectrum deal
- SDxCentral: SpaceX teases bold Starlink mobile plans
- Kajeet: A guide to cellular frequency bands used by US carriers
- Dgtl Infra: 5G spectrum explained, low, mid and high band
About this article: GeekBlog covers U.S. technology news, AI, phones, smartwatches and gaming. Every story is written and checked under our Editorial Policy. Spotted a mistake or have a story tip? Contact our editors.
