On October 29, the FCC voted 3–0 to propose letting satellite operators reach ordinary phones without renting spectrum from the phone companies, and the phone companies had already filed, a week before the vote, to strip the two biggest bands out of the proposal. Nothing was granted. It is only a proposal, only a comment period, only 482 megahertz of eligibility. Just the tip. Everybody in the docket knows what the tip is for. The 2024 rules made AT&T, Verizon, and T-Mobile the landlords of every satellite conversation with a normal phone, and the new proposal retires the landlords. The carriers will survive the satellite business. The rent is what dies, and the rent is the part they filed to protect.
The source video is Joseph Cristina at DISHYtech, Starlink Mobile Just Won! The Big 3 Are Done. His headline runs ahead of the record, and the record, checked against the FCC's own documents, is more interesting than the headline.
What the FCC actually did
The October 29 vote adopted a Notice of Proposed Rulemaking, FCC-26-71, unanimously. It proposes expanding the Supplemental Coverage from Space framework from five bands and 344 megahertz to thirteen bands and 826, adding 482 megahertz of eligibility, including the 2.5 GHz and 3.45 GHz mid-bands. It proposes harmonizing the satellite rules. And it proposes replacing the 2024 framework's central mechanism: today a satellite operator may work in a terrestrial band only by leasing spectrum from the terrestrial licensees who hold that band, all of them, across a whole geographic area. The proposal swaps the lease for a certification, the operator attesting that it has coexistence agreements or holds the licenses itself, in smaller service areas. Comments run into January. Final rules need a second vote, sometime in 2027. Commissioner Gomez voted yes and said on the record that she does not want the framework to let one or two companies dominate the market. She is tracking the right variable.
Three weeks earlier, on October 6, the FCC's bureaus had already granted SpaceX the working prototype of the new model: DA 26-1078, a license for 15,000 direct-to-cell satellites, with the lease requirement waived for the bands SpaceX bought outright from EchoStar for roughly $19.6 billion. The license is real and it is conditioned: the EchoStar bands stay test-only until that transaction finishes closing, expected around November 30, 2027, half the fleet is due by October 2032, all of it by October 2035, and not one of the 15,000 has launched. Authorization is the starting gun. The video sells it as the finish line, and it is the starting gun.
The petition, honestly
The carriers' move was a joint ex parte letter from their trade association, CTIA, filed October 22 after meetings with all three commissioners' offices on October 20. It asked the Commission to remove the 2.5 and 3.45 GHz bands, 290 of the 482 megahertz, from the proposal before the vote. T-Mobile filed separately on October 21 with the technical case: terrestrial 5G in those bands runs synchronized time-division, every tower transmitting and listening on one shared clock, and a satellite's signal arrives late and Doppler-shifted from a platform moving seven and a half kilometers a second, which can drop downlink energy into the slots where towers are listening. The interference mechanism is real engineering. T-Mobile's magnitudes, up to 96 percent of a cell's uplink coverage lost, are T-Mobile's modeling in an advocacy filing, and the FCC has adopted none of them. The Commission kept both bands in the proposal and sent the argument to the comment round. The carriers lost the ambush and kept the war.
Just the tip
There is an old joke about a negotiation conducted one inch at a time, and every regulated industry in America knows it by heart, because every regulated industry has run it. Just a proposal. Just a comment period. Just an eligibility list. The party being entered learns the truth at the pace the process allows, and the process is designed so that each individual step is too small to justify a war. The carriers have used this exact instrument for forty years, one waiver, one ruling, one leased band at a time. They recognized the shape of October 29 instantly, the way you recognize your own handwriting. That is why the filing came before the vote, why it aimed at 60 percent of the megahertz, and why it argued architecture instead of price. You do not mobilize three legal departments over a tip. You mobilize over what the tip is attached to.
The velodrome
Track sprint racing works like this. Two riders on a velodrome go super slow, super slow, super slow, sometimes stopping dead on the banking, because the rider in front does all the work and the rider behind rides free in the slipstream. The race moves at the speed of the man who refuses to lead. Then one rider decides he has the hubris to make a break for it, and in that second the race exists, and everyone on the track has to chase or lose. Incumbent industries are a velodrome. There is no reason to innovate except to chase, and for thirty years the American phone business rode in a circle at walking pace, charging exorbitant fees for the privilege, on networks the rest of the world had already lapped. Musk's entire method is the break. If you cannot win the red tape war on the terrestrial world, leave the terrestrial world. Electric cars, tunnels instead of highways, nerve repair, rockets, and now the phone network itself: he takes off first, in public, at his own expense, and an industry that swore the money did not exist discovers an innovation budget by Friday.
I watched this movie in the nineties
In the 1990s, Japan was mocking us, and mocking Europe too, because Japanese phone networks had early versions of everything while Americans paid premium prices for systems that were not even GSM. I was in England in 1990, when a mobile phone was a rumor. I came back in 1996 and everybody had a Nokia or an Ericsson in hand and the whole country was obsessed with SMS, doing casually in 1996 what Americans would not normalize until BlackBerry, years later. In between, on a catamaran Mark Harrison and I piloted from Acapulco to Los Angeles, I tried desperately to text my girlfriend Wendy, and the answer from America was unanimous: nobody liked texts, nobody wanted texts, leave me alone. The American carriers sat still for a decade while the world lapped them, and they were paid exorbitantly for sitting. I am not guessing what a comfortable incumbent does with a quiet decade. I have the receipts, in the form of my own phone bills, from three countries.
The physics, without the thumb on the scale
Shannon's theorem caps what one channel can carry: bandwidth times a factor of signal quality. A satellite beam spreads one channel over a footprint that can reach thousands of square miles, shared by everyone standing in it. A terrestrial network reuses the same frequencies in thousands of small cells, a few square miles each, and multiplies its capacity by the count of cells. Geometry therefore settles the city argument permanently, and the link budgets settle the indoor argument: a phone's whisper of a transmitter, hundreds of miles below a satellite, behind a roof, generally does not close the link. This is why SpaceX's own checkbook describes a hybrid future, $19.6 billion of purchased terrestrial spectrum, an 800 megahertz deal reported near $8 billion more, because low bands travel and penetrate. The Big Three are not done as networks, and no honest reading of the physics says so. What is done, if the proposal survives its comment period, is their position as the toll every satellite phone call must pay. Note the control experiment the FCC ran in April: AST SpaceMobile's satellite service was approved under the old lease model, carriers as landlords, without a fight. Satellite service was never the threat. Satellite service that owes them nothing is the threat.
What happens next
The comment record runs into January 2027 and final rules need a second vote. SpaceX plans to start launching its second-generation mobile satellites in 2027, states service on its own spectrum by the end of 2027 or early 2028, and needs new phone chipsets, arriving across 2027 and 2028, before the new bands work in anyone's pocket. Musk, answering Verizon's estimate of four to six years and $300–500 billion for a competing network, said two or three years. Your phone bill does not change this cycle. What changed on October 29 is narrower and larger: the federal government proposed, unanimously, to stop requiring the future to rent the past. The last paragraph of every story like this one is the same, so here it is plainly. The carriers were never beaten by a satellite. They are being repealed as landlords, one proposal at a time, just the tip first, and they know exactly how the rest of the joke goes.
House note, chrisabraham.net being the laboratory: the facts above were checked against the FCC record (the October 29 NPRM, the October 6 authorization, and the carriers' October filings) before this post was written. Where the source video's framing outran that record, the post says so and prints the record.