Apple Bet on a Folding Screen. SpaceX Just Got 15,000 Satellites.

Apple spent this month changing the shape of the phone. SpaceX spent it changing what the phone connects to.

Apple spent this month changing the shape of the phone. SpaceX spent it changing what the phone connects to.

On October 6, the Federal Communications Commission granted SpaceX permission for up to 15,000 direct-to-device satellites, spacecraft designed to talk straight to ordinary phones. The satellites will fly between 326 and 335 kilometers up. The grant is partial and comes with conditions, but it includes a waiver so SpaceX does not need a lease deal with a ground-based carrier.^1, 2

Half the satellites must be working by October 2032 and all of them by October 2035.³ SpaceX says the new system will deliver 150 megabits per second per user. Today’s satellite texting service runs on about 650 satellites at roughly 4 megabits.¹

A phone that does not need a tower

That is the part that matters. For decades, a mobile phone has been only as good as the nearest cell tower. If a satellite network can deliver real broadband to an ordinary handset, the tower stops being the gatekeeper. Coverage would no longer end at the edge of town, the edge of the highway or the edge of the country.

SpaceX has been assembling the pieces for some time. In September 2025 it agreed to buy spectrum licenses from EchoStar for about $17 billion, paid partly in cash and partly in SpaceX stock.⁴ It has launched nearly 13,000 Starlink satellites, about 11,150 of which are still in orbit, and together with a grant in January its approvals now cover 30,000 satellites.² Each new satellite adds capacity to a network that is being designed from the start to talk to phones, not to dishes on rooftops.

Consumers are paying attention. In a TD Cowen survey, 34% said they would consider switching to a SpaceX mobile service.³ That is a striking number for a service that does not yet exist in full. It suggests a meaningful slice of the market is not especially loyal to its current carrier, and would move for coverage that works everywhere.

✓ Trusted partner A former Wall Street tech trader who says he saw the smartphone boom coming early has now named the one stock he believes is best placed if a satellite-connected AI phone arrives →

The carriers say relax

The incumbents are not worried, or at least say they are not. In September, T-Mobile’s chief financial officer said Starlink “is not a competitive threat from a wireless perspective at all,” and that “the physics just don’t allow direct-to-cell to be a substitute product.” T-Mobile has said its satellite service accounts for 0.0002% of its network usage. Verizon’s chief executive said he had seen “zero discernible impact” on broadband growth from satellite providers.⁵

That may prove right. It is also exactly what established players tend to say in the early innings of a technology that later reshapes their industry. The question for investors is not what the carriers say today. It is what the network looks like when thousands of new satellites are in orbit.

Physics is a real constraint. A satellite hundreds of kilometers up has a harder job than a tower down the street. But SpaceX has spent more than two decades engineering around physics, and the new satellites will fly lower, at roughly 330 kilometers, to close the distance.¹ None of this is guaranteed: the full network will take years to build, and the FCC grant comes with conditions.³

Apple’s bet looks very different

Apple’s big move this year went the other way. Its first foldable, the iPhone Duo, starts at $1,999. Pre-orders opened October 16 and the phone launched October 23. Demand was strong enough that shipping dates for new orders slipped to January 29, 2027.^6, 7

Reviewers were less sure. Their main complaint was weight, at 254 grams, and they also knocked the switch to Touch ID, the lack of a telephoto lens and heat. One major tech site called it one of Apple’s messiest first-generation products in a long time.⁸ Engadget did find the battery strong, measuring almost 29 hours of video playback.⁸

A foldable is a real innovation, and it may sell very well. But it is still a phone that depends on the same towers, the same carriers and the same model of how a phone works. It changes the hardware. It does not change the network.

✓ Trusted partner See the research behind this story →

So is Musk building a phone?

Officially, no. In July, after the Wall Street Journal reported that an AI handset prototype had been shown to investors before SpaceX’s listing, Elon Musk called the reports “utterly false.” A year earlier he had said, “if we have to make a phone, we will, but we will aspire not to make a phone.”^9, 10

Neither statement rules the idea out. Both are statements of leverage. A company with the satellites, the spectrum and an AI platform does not need to announce a phone to change the phone business. It only needs the network to be good enough that someone, maybe SpaceX, maybe a partner, puts it in a device people want. Musk has also made a habit of entering industries that incumbents assumed were locked up, from electric cars to reusable rockets.

Where the money gets made

The biggest fortunes from the smartphone era were not made by people who bought the obvious name after it was obvious. They were made by those who saw the shift coming and identified the companies the new devices could not be built without.

The company with the most famous phone is rarely the only winner. Every major device shift has created a second tier of suppliers, the makers of chips, radios, screens and batteries, whose sales rose with every unit sold.

The same is likely to be true here. A satellite-connected AI phone, if it comes, will need chips, antennas, components and software from suppliers most investors have never heard of. Those are the companies that can move the most when the market finally catches on.