Elon's Insane SpaceX Phone Announcement

Picture a hiking trail. No bars. You are standing outside under open sky, you type a text, and it goes through anyway because your phone is talking to Starlink. That is how most people think of Starlink on a phone. A backup.

Then you walk inside a building, and you lose it.

AT&T's CEO, John Stankey, put it this way in September: "It doesn't get through buildings very well. It does very well line of sight."

But in that same answer, he added a catch. If you have lower band spectrum, he said, then it will have some decent penetration into some buildings.

About 4 weeks later, on October 8th, SpaceX agreed to buy low band spectrum across the entire country. About 20 minutes after that, Elon posted: "This is the last critical piece of the spectrum puzzle needed for SpaceX to provide complete phone coverage in America." In after-hours trading, AT&T, Verizon and T-Mobile shares fell as much as about 5 to 7.5%.

My take: SpaceX just checkmated an entire industry. Here is why. One note first: nothing here is investment or financial advice.

Two pieces of news

This week brought two pieces of news, both about the phone service SpaceX calls Starlink Mobile.

First, on October 6th, the FCC approved SpaceX's plan for 15,000 new satellites built to talk straight to regular phones. The FCC is the government agency that decides who gets to use which airwaves. That approval is done.

Second, on October 8th, SpaceX agreed to buy a nationwide slice of airwaves in the 800 MHz band from a private investment firm called Grain Management. That one is not done. The FCC still has to approve the sale.

MHz counts how many million times a radio wave wiggles every second. That number has a lot to do with whether a signal gets through your walls.

Bass and treble

To see why Elon called this the last piece, start with spectrum. Every phone signal is a radio wave, and spectrum is the full range of those waves, sorted by how fast they wiggle. Low band is the slow end, with long waves that travel far. Mid band wiggles faster. Its waves are shorter and carry a lot more data, but they don't travel as far.

Think about your neighbor playing music on the other side of your wall. You hear the bass just fine, those low deep notes. The high notes, the treble, mostly get soaked up by the wall before they reach you.

Radio waves behave the same way. At 800 MHz, each wave is about 15 inches long. At 2 GHz, which is 2,000 MHz, each wave is only about 6 inches long. So 800 MHz is the bass and 2 GHz is the treble. The longer wave bends around corners and loses less strength going through walls and windows. SpaceX says the 800 MHz band has superior building penetration.

The bass comes with a trade-off. Through the wall, you can feel the beat, but you can't make out the words. Low band reaches far and gets inside, but it carries less data. For phones, mid band carries the speed. A phone network that works everywhere needs both: the bass to reach you and the treble to move your data fast.

A backup today

So what does Starlink on a phone do right now? In the US, T-Mobile sells it as T-Satellite for $10 a month. Satellites talk straight to the normal phone in your hand. You can send texts and picture messages and use a handful of apps, but it needs you outdoors with a clear view of the sky.

It also runs on a thin slice of treble. SpaceX's Gwynne Shotwell said the current system uses about 5 MHz, and it runs through local phone companies.

That is fine for dead zones, and that is how AT&T sees satellites. Stankey says AT&T already meets about 98% of the use cases, and satellite is for that 2%. But in the same interview, he gave a number that points the other way. In urban Chicago in the middle of the day, over 40% of the traffic is coming from inside buildings.

A network that can't reach you indoors can't be your main carrier, the company you pay every month, because so much of your phone use happens inside. That is why I think the walls decide whether Starlink stays a backup or becomes a phone company people actually switch to.

The last piece

Now put the puzzle together, because SpaceX has been collecting the pieces for years.

The first piece is the satellites. SpaceX has been launching phone satellites since January 2024, and it says the 15,000 next-generation satellites the FCC just approved will have more than 100 times the bandwidth of the current generation. Bandwidth is how much data can flow at once. That is the jump from mostly texting to actual high-speed service on a regular phone.

The second piece is the treble. In September 2025, SpaceX agreed to buy mid-band spectrum from EchoStar, with the main deal worth about $17 billion. A November 2025 add-on of more airwaves brought it to 65 MHz and about $19.6 billion in total, and the FCC approved the transfer in May 2026. Shotwell put the EchoStar spectrum at 65 MHz, compared to the roughly 5 MHz the system uses now. And this week, the FCC gave SpaceX a waiver so it can use those airwaves for satellite-to-phone service without a phone company as its partner. Until the EchoStar purchase closes, which SpaceX expects in November 2027, it can only test on those airwaves.

The bass was still missing. That is the 800 MHz deal. It is up to 14 MHz of paired spectrum, which means one part for sending and one part for receiving, and it covers the entire country. The FCC itself called this block "a valuable source of paired, broadband-ready, low-band spectrum" that has "been underutilized in recent years." It sat mostly unused, and SpaceX agreed to buy all of it.

That is the moment Elon's post was about. It came about 20 minutes after SpaceX's announcement, quoting a post about the AT&T walls argument. Two minutes later, he posted again: "Very big deal."

Line up the pieces and you can see why he picked the word puzzle. SpaceX already had the satellites and the treble. Now it is buying the bass, the piece that gets inside.

The high-rise problem

There is a fair point on the other side, and Stankey made it himself. Even with lower band spectrum, he said, "it's not going to penetrate the skin of a high-rise." And a satellite is a long way up.

Think about your neighbor's music again. The bass comes through loud and clear when the speaker is right next door. Put that speaker a mile away, and you hear a lot less of it.

That is why SpaceX plans to use the ground, too. It says it will put the 800 MHz band on its own ground radios on towers, rooftops and other structures, and it is filing to add the same band to its satellites. So where a building is in the way, the speaker moves next door. Where there are no towers at all, the satellites cover it from above.

SpaceX calls it one network that provides Americans with a reliable service "indoors, outdoors, in cellular dead zones, and everywhere in between." It also says most existing mobile devices already support this underused band. For most people, that means no new phone.

What still has to happen

None of this new network is live yet. The FCC has to approve the sale of these licenses, and its own goal for reviews like this is about 180 days, roughly 6 months once the review formally starts. The price was not disclosed. The next-generation satellites are not in orbit yet.

In August, Shotwell said, "We will start to fly the satellites next year, and we will start providing service end of next year." So her target is service by the end of 2027.

My rule of thumb with any Elon company is simple. The direction is almost always right, and the timeline is almost always late. So stop counting misses and count deliveries.

On phones, SpaceX has already delivered once. Its first phone satellites went up on January 2nd, 2024, and the first texts went through them 6 days later.

A giant pie

So why is this such a big deal? Start with the phone companies: AT&T, Verizon and T-Mobile. In August, Shotwell said those three make roughly $600 billion a year between them. Their own reports add up to about $350 billion for 2025, and that's still a giant pie. And she said, "I anticipate us to be able to acquire quite a few of their customers because I think our service will be better."

Then look at the timing. The announcement came out 19 minutes after the stock market closed, and all three stocks had closed up that day. My read is that the after-hours drop was the market reacting to this deal. The carriers have already started to respond. On October 1st, they made a joint venture official to pool their own spectrum for satellite-to-phone service.

For a regular person, this is where it gets good. When SpaceX first announced phone service with T-Mobile in 2022, it said more than 20% of the land area in the US had no cell service from the ground. And SpaceX says more than 1.5 million people were able to communicate through its satellite service when cell networks went down.

Now put that same service inside your house. Even if you never switch, I think a new competitor that owns its whole network pushes every carrier to get better and cheaper.

The phone is next

So where does all of this lead? SpaceX has the satellites, and now it is buying the airwaves to go with them. The only piece left is the phone itself. My bet is that SpaceX makes its own phone. Hook that thing up to Starlink, and it's game over.

Elon has already laid out where this goes: if growth continues, Starlink will one day carry the majority of internet traffic. At that point, it is the internet, and everything else just connects to Starlink.

Now go back to what AT&T's CEO said in September. The satellite signal doesn't get through buildings very well. He was right about that. But in the same breath, he named the fix: lower band spectrum. About 4 weeks later, SpaceX agreed to buy it across the entire country.

Back on that hiking trail, your text goes out under open sky. Then you drive home and walk inside. If SpaceX pulls this off, the signal comes with you. That bass is what turns Starlink from a backup on a hiking trail into a phone network that reaches you inside your own home.