FCC Chairman Brendan Carr just did something unusual…
Amazon (AMZN) had just filed a 17-page petition with the Federal Communications Commission (“FCC”), urging the agency to deny SpaceX’s application to deploy up to 1 million satellites in low Earth orbit. The petition called it “a lofty ambition rather than a real plan,” according to Bloomberg.
Amazon’s argument went something like this: Deploying a million satellites would take more than 220 years at the current pace of global launches. Amazon also warned that approving SpaceX’s application would give the company near-total control over insertion orbits, forcing every other operator to plan around a constellation that might never exist.
Carr wasn’t having it. He publicly fired back at Amazon by name on X, the social media platform formerly known as Twitter. And he didn’t mince words…
Amazon should focus on the fact that it will fall roughly 1,000 satellites short of meeting its upcoming deployment milestone, rather than spending their time and resources filing petitions against companies that are putting thousands of satellites in orbit.
Carr is referencing Amazon’s upcoming July 30 deadline to have half of its planned 3,232-satellite constellation deployed. It currently only has 212 satellites in orbit. That’s a massive gap of 1,400 satellites that Amazon needs to launch in less than five months.
Then Carr twisted the knife in his remarks to Reuters:
“Given the pace at which Amazon is launching satellites, I can see why they would think it would take other people centuries to launch,” Carr said. “I don’t anticipate the Amazon filing will get much traction here.”
It was a bad day for Amazon’s satellite ambitions. And it was a revealing one for anyone watching where the space industry is headed.
What SpaceX Is Working Toward
SpaceX isn’t just asking for permission to launch a million internet satellites…
It wants to build data centers in orbit.
If that sounds like science fiction… solar-powered satellites running artificial intelligence (“AI”) agents, linked together with optical connections… that’s because just a few years ago, it would have been.
But SpaceX filed this application on January 30, just days before it completed a $1.25 trillion merger with xAI, Elon Musk’s AI company. An internal SpaceX memo obtained by the New York Times called the combined entity “the most ambitious, vertically-integrated innovation engine on (and off) Earth, with A.I., rockets, space-based internet, direct-to-mobile device communications and the world’s foremost real-time information and free speech platform.”
Ars Technica had more details from the memo…
Musk told employees that launching 1 million tons per year of satellites, generating 100 [kilowatts] of compute power per ton, would add 100 gigawatts of AI compute capacity annually, “with no ongoing operational or maintenance needs.” Ultimately, Musk believes there is a path to launching 1 [terawatt]/year from Earth.
“My estimate is that within 2 to 3 years, the lowest cost way to generate AI compute will be in space,” Musk wrote. “This cost-efficiency alone will enable innovative companies to forge ahead in training their AI models and processing data at unprecedented speeds and scales, accelerating breakthroughs in our understanding of physics and invention of technologies to benefit humanity.”
These claims are not a sure thing, to be clear. Deutsche Bank has noted that “there are clearly technical challenges to making this a viable endeavor, but these seem to be engineering constraints as opposed to physics.”
Nonetheless, Deutsche Bank says that orbital data centers could “reach close to parity” with traditional data centers built on Earth as soon as the 2030s.
And of course, Musk has a track record of moving far faster than experts predict…
I recently wrote about Musk’s “Project Apex” supercomputer in Memphis and the broader AI arms race it represents. The orbital-data-center plan is a clear extension of that story.
xAI built the most powerful AI training cluster on the globe in a matter of months… Now Musk wants to take that same approach to space.
And investors are paying attention.
The SpaceX IPO Everyone Wants In On
In the coming months, we’ll likely see the largest initial public offering (“IPO”) in American history…
I covered how investors can access the combined SpaceX/xAI IPO, including some newer financial vehicles that let everyday investors buy in before it goes public.
And the numbers behind the valuation tell an incredible story. As I noted…
SpaceX has more than 9,000 satellites in orbit and adds roughly 60 more with every mission… And because SpaceX reuses its own rockets, launch costs have plummeted more than 90% compared with the old model of building a rocket, firing it once, and letting it sink into the ocean.
So far this year, the company has launched more than 600 Starlink satellites… and the FCC approved SpaceX’s previous request to expand its Starlink constellation to 15,000 satellites back in January.
Compare that to Amazon…
Jeff Bezos’ company has invested more than $10 billion in its competing Project Kuiper satellite service, according to CNBC. But as we noted earlier, it has only a few hundred satellites in orbit. And Amazon even recently asked the FCC for a 24-month extension on its deployment milestone, blaming rocket shortages and manufacturing disruptions.
Meanwhile, you can track SpaceX’s upcoming missions in real time online… including two that are expected to launch tomorrow from Florida and California. You can even dig into a wild amount of detail on its more than 600 completed missions.
And SpaceX’s next chapter is about to start…
Musk just announced that the first flight of Starship V3, the most powerful rocket in SpaceX’s lineup, is roughly four weeks away. As Space.com reported:
The V3 upgrade makes significant changes to Starship’s Super Heavy booster and Ship upper stage. Both are slightly taller than their predecessors and produce stronger, more efficient thrust using SpaceX’s upgraded Raptor 3 rocket engine. Previous Starship iterations were already the biggest and most powerful rockets ever to launch, and V3 will continue to up the ante.
If Starship V3 works, it will be the rocket that makes the million-satellite constellation at least theoretically possible. It will also serve as NASA’s lunar lander for the Artemis 3 mission, now scheduled for 2027.
As I mentioned last week, there are several ways that interested investors can get access to SpaceX before it goes public…
One of the most direct proxies is EchoStar (SATS). It sold wireless spectrum licenses to SpaceX in exchange for SpaceX stock a few years back.
At one point, EchoStar was even trading at a discount to its SpaceX holdings. That’s no longer true today, but buying shares of EchoStar does remain as simple as making any stock trade… and would give you exposure to SpaceX.
There’s also Alphabet (GOOGL), which reportedly owns about 7.5% of SpaceX after an investment years ago. Still, Alphabet has a $3.6 trillion market cap… So even if SpaceX goes public at $1.75 trillion, Alphabet’s stake would be worth around $130 billion. That’s significant… but it won’t be the biggest mover of GOOGL stock.
And my colleague Jeff Brown at Brownstone Research has shared a special way for everyday folks to claim a more direct stake in this private venture before its potential IPO… and you can start with as little as $500.
Out of respect for his paid subscribers, I can’t share that specific investment here. But his research service, called The Near Future Report, provides specific details on which vehicles he recommends and simple step-by-step instructions for buying into SpaceX pre-IPO shares, much the way you’d buy any stock.
Of course, if you’re interested in getting access to The Near Future Report, you can click here to go directly to a subscription order form – please note, this link does not go to a long video.
What This All Means for Investors
So where does this leave you?
First, SpaceX is almost certainly going public this year…
The company has hired Bank of America (BAC), Goldman Sachs (GS), JPMorgan Chase (JPM), and Morgan Stanley (MS) to lead the deal. That’s essentially every major Wall Street bank.
The question is valuation, timing, and whether the narrative on orbital data centers holds up to scrutiny from public-market investors.
Second, the AI infrastructure build-out is accelerating in ways that most folks haven’t fully processed. And frankly, most won’t until it’s too late.
SpaceX merging with xAI, building orbital data centers, planning Starship V3 launches every few weeks… all of this is happening against a backdrop of unprecedented spending on AI computing power.
As I wrote in the Project Apex piece, Goldman Sachs estimates that AI could add $7 trillion to global GDP over the coming decade. And Morgan Stanley sees a $40 trillion total addressable market for AI technology.
Of course, that doesn’t mean that everything will be rosy for America… especially American workers.
As I speculated in late February, AI could ultimately usher in a “doom loop” for jobs and the stock market. If and when that happens, I personally want to own the companies behind the AI and its energy demands.
And that brings me to my last point… The energy demands of AI are reshaping entire sectors. Data centers already consume 4.4% of total U.S. electricity, and that number is climbing fast.
That’s a big reason why Musk is so focused on putting data centers in space. It’s also why nuclear energy stocks have been surging. And it’s why investors should be looking at the infrastructure companies supporting this build-out, not just the headline names like Nvidia (NVDA).
Amazon’s failed attempt to slow down SpaceX at the FCC tells you something about the competitive dynamics here. When your main strategy is to petition regulators to stop your competitor, it usually means you’re losing.
And Elon Musk is making his biggest bet yet.
Whether he’s right or not, you’ll want to be paying attention. For more on the AI investment landscape, including how to access a free stock recommendation from growth investor Louis Navellier, check out our recent deep dive on Project Apex and why he’s calling it a potential “ChatGPT Killer.”
