Today on Decoder, I’m talking to Ryan Mac, a technology reporter at The New York Times and coauthor of the excellent book Character Limit: How Elon Musk Destroyed Twitter, which came out in 2024. I can’t recommend it enough.
I wanted to have Ryan on the show because we’re on the cusp of the SpaceX IPO, which promises to be one of the most consequential public offerings in history for a variety of reasons — its biggest-ever size, of course, at nearly $2 trillion dollars, but also because all kinds of rules that keep our markets fair are being bent, if not outright broken, along the way. I also wanted to talk to Ryan because buried somewhere inside SpaceX is X, the social platform formerly known as Twitter, which Musk purchased in 2022. That’s what Ryan cowrote that book about.
I was very confident that Musk would come to regret buying Twitter back then. I wrote a piece called “Welcome To hell, Elon,” which is probably the single most-read thing I’ve ever written. My thesis was that there would be no way to grow Twitter users and revenue without moderating the platform well and, ultimately, that Elon buying Twitter would destroy his reputation and cause damage to his other companies.
Now, we have the numbers from the SpaceX IPO filing to see how right my prediction was. X is shrinking by every major metric, but it may not matter, as Ryan points out. Take a listen, and let me know what you think.
Ryan and I also got into all those rules being broken to land the SpaceX IPO — rules about shareholder control, inclusion in the major index funds, and all the other levers of market accountability that usually serve to keep companies in check. You’re going to hear us say “corporate governance” a lot in this episode, and while it may sound boring, it won’t be if you take a shot every time it comes up.
Okay, don’t do that. But do consider what it means that Elon has become so rich, so powerful, and so detached from the levers of accountability that he can apparently get away with anything. That’s all without any of the major fund managers or investors calling foul because they don’t want to miss out on what could be the biggest financial windfall in recent memory. There’s a lot to think about in this episode.
Okay: New York Times tech reporter Ryan Mac, on Elon Musk, X, and the SpaceX IPO. Here we go.
This interview has been lightly edited for length and clarity.
Ryan Mac, you’re a technology reporter at the New York Times. Welcome to Decoder.
Thanks for having me.
I am really excited to talk to you. I can’t believe you’ve never been on the show before. I feel like we’ve done a lot of reporting in and around each other. I’m a big fan. Thanks so much for being on.
I know. What the hell, man? You just have avoided me this whole time. But no, I’m kidding. It’s good to be here. I’ve listened to many episodes, so great to be a part of it.
Today on Decoder, I’m talking to Ryan Mac, a technology reporter at The New York Times and coauthor of the excellent book Character Limit: How Elon Musk Destroyed Twitter, which came out in 2024. I can’t recommend it enough.
I wanted to have Ryan on the show because we’re on the cusp of the SpaceX IPO, which promises to be one of the most consequential public offerings in history for a variety of reasons — its biggest-ever size, of course, at nearly $2 trillion dollars, but also because all kinds of rules that keep our markets fair are being bent, if not outright broken, along the way. I also wanted to talk to Ryan because buried somewhere inside SpaceX is X, the social platform formerly known as Twitter, which Musk purchased in 2022. That’s what Ryan cowrote that book about.
Well, now we’re going to ask you to answer for your crimes, which is what the Decoder audience really wants me to do, I guess.
Speaking of crimes, we’re going to talk about the SpaceX IPO. Elon Musk has obviously filed to take SpaceX public. There’s a lot in that IPO, including the idea that there’s a $28 trillion addressable market for SpaceX services, which is more than the world. Just a lot in there.
You’ve reported on a lot of it. So I do want to dive into it, but I actually want to start with X, the everything app, the app formerly known as Twitter. Because the SpaceX S-1 really gives us our first look into what that business is, what it has become, where it’s growing.
In 2022, I wrote an article — maybe the most viral article I’ve ever written — it was called “Welcome to hell, Elon,” in which I very confidently predicted that buying Twitter would be a disaster for Elon Musk. I’m just going to read you my thesis. It was the first sentence of the piece. And then I want to try to back into what we know about X. I’m very curious if you think this has come true or not.
So my thesis was: “Twitter is a disaster clown car company that is successful despite itself and there is no possible way to grow users and revenue without making a series of enormous compromises that will ultimately destroy Elon Musk’s reputation and possibly cause grievous damage to his other companies.”
There’s one view to say, “Yep, that totally came true.” There’s another view to say that actually Elon is more powerful than ever and on the cusp of an IPO that’s going to make him a trillionaire. So tell me about X. What do we know about X, the company in the years since Elon has bought it and what do we know about its financials as reported in this S-1?
Sure. I think you mentioned the word “growing” in all that and I think the place to start is the fact that X is simply not growing. It’s stagnated in terms of revenue, stagnated in terms of user growth. It’s been buried twice within Elon’s companies — first into xAI and now into SpaceX. So it’s become, in some ways, an afterthought in the Musk empire, despite it still being arguably Musk’s favorite thing. He spends countless hours a day on that thing, like many of us used to, and many of us still do.
But in terms of a business proposition, it’s a non-factor if you compare it to some of the other aspects of this business — something like Starlink, for example. If you look back at 2022, it’s just bizarre. He bought this company on a whim. He pitched this idea to investors that he would have one billion users. He would have integrated payments. It would be somewhere you could potentially book a taxi. He pitched this idea of it being WeChat. You mentioned the everything app. And it’s certainly not the everything app. At one point he was like, “You could watch TV on it.”
None of that has come to fruition. Yet I look at what’s happened in the last four to five years since then, and he’s gotten more powerful than ever. His net worth has increased. I think around the time he bought the company he was around $300 billion. His net worth now fluctuates anywhere from $600 to $800 billion these days. And a SpaceX IPO will take him potentially beyond the trillion dollar mark for the first time ever in human history. So it’s bizarre in that there are a lot of contradictory things about it, but at the end of the day, I’d argue he still comes out on top.
Is it just as simple as he bought a distribution platform for his own tweets and he controlled it and he fixed the algorithm to favor himself and that worked? And it doesn’t matter that revenue is down $100 million year over year and not even quite 40 percent of Twitter’s pre-acquisition revenue?
He’s destroyed the business by every metric we can see in the S-1. Every number is down. And only the revenue from data licensing to AI companies is up.
His own AI company too. Yes, if you singularly look at X as a business, it’s clearly a failure from the time he took over the company to Fidelity marking the valuation of the company down to $10 billion before he merged it with xAI. But also you have to look at it in the whole landscape of Musk Inc. Since he bought the company, he spun up xAI, raised billions of dollars for that company. He then merged it into xAI, burying it, and then he merged it again with xAI into SpaceX.
I guess he’s up, if you’re doing a plus-minus analysis of valuations of these companies. Again, these are valuations that seemingly have no basis in business fundamentals. We’re playing with Musk math here. He has this whole cadre of investors and friends that are willing to back him to the end of the Earth, but yeah, he’s, I’d say, winning.
There’s a version of this where you could straightforwardly make the argument that however many billions he lost on Twitter is worth it as an investment that got him to, “We’re going to do a trillion-dollar SpaceX IPO.”
Yeah, and I caution against looking back at it historically and thinking that was his plan all along. There is this trope that came up after Trump won the election that Elon Musk bought X to then help elect Donald Trump. There was really no proof of that, especially when we did our reporting at The Times and for our book Character Limit, but it has worked out for him. It’s indisputable in that sense.
He has bought a distribution platform for his own tweets. He’s the most followed person on the platform now. He controls the algorithm, he controls the content that gets boosted on the platform. I don’t know how to say this more, but he’s winning and that’s just where we are right now in society.
In the pre-X days when it was still Twitter, Elon would constantly talk about how he didn’t do any marketing for Tesla. They had spent no money in advertising, no money in marketing. He would just tweet to move Tesla sales up and down. And there was infinite demand for the Tesla Model 3 in particular and the Model Y. So much so that every other car maker essentially got confused and made Tesla Model Ys of their own.
They just thought people wanted electric cars. And maybe they just wanted Tesla’s and maybe there was a meme stock component to it, but Elon was just very good at using Twitter to drive demand for his products. He would constantly say, “I don’t have to pay for marketing. I just have this platform.”
The other part of my thesis was that buying X and changing the algorithm and being as political as he has been would cause reputational damage to Elon, would cause reputational damage to his companies. There’s some evidence that that is true in the case of Tesla, where the cars are less popular than they once were, certainly. There are campaigns to protest at Tesla dealerships. Is that true for SpaceX? Is it true for the rest of his empire?
It’s a great question. And with Tesla, of course, after the election, you saw the Tesla Takedown protests, people slapping those bumper stickers on their cars: “I bought this before Elon went crazy.” You could see the share of the Tesla market in the EV market falling. Of course, there’s a lot to do with the rise of Chinese vehicle manufacturers like BYD.
But I don’t know. I don’t think I see the same kind of reputational harm to SpaceX that I’ve seen with Tesla. And that may be because there’s just not as much consumer contact with parts of SpaceX. I think of something like the launch business. I’m not going out and buying a launch. Regular people aren’t buying launches; SpaceX is contracting with governments and big companies. Not to mention that they largely have a monopoly on getting things into space.
If your option is working with a company that has a CEO that is reputationally compromised versus not getting into space at all, you’ll probably go with the former. Starlink is another basket. It offers a product that is quite good and is not challenged in any way. I think of something like Iridium, for example, or the companies it competes against, and it just blows them out of the water. It’s such a strong service, so much so that governments rely on it and Ukraine relies on it and the numbers there continue to grow. That’s the crown jewel of the SpaceX business empire right now in terms of revenue and profit.
There’s no moral case to be made for being on Hughesnet still. You can’t be a better person by being on some of the other satellite providers that service the rural parts of the country. We see it in our own traffic and our own comments. We cover Starlink. There’s nothing better. They continue to innovate in the ways they continue to innovate and the audience doesn’t like it, but then there’s a huge part of the audience that says, “Wait, I don’t have a market alternative to this.”
You’re saying where there isn’t a market alternative, the reputational issues have not been a problem and every place where there might be or there’s a consumer market, the reputational issues have damaged it. I think X is actually the example of this. There are a number of market alternatives to X, so people have just left the platform.
Whether or not there’s one big competitor to X remains to be seen. Threads by some accounts is vastly bigger than X, but it does not have the influence. Bluesky is run by very charming, very ideological people. They’re doing whatever it is they’re doing. It’s certainly not a competitor head up to X. Is there a reason that influence hasn’t recapitulated itself anywhere, or that the people still on X have stayed there?
Oh man, it’s something I think about a lot and I think the rumors of X’s demise at the time were greatly exaggerated in a way. These social platforms are so sticky. I remember one of the rounds of layoffs — I think it was the “A Fork in the Road” or one of those earlier rounds — and Twitter went down for a period of time, too. I remember being at dinner and everyone just writing eulogies for Twitter that night. And I’m like, “Guys, I don’t think this is it.”
This is a pretty resilient platform. It’s been developed over more than a decade. It’s been around for a while. It’s quite a resilient thing and it has a dedicated user base. That’s why we love it. It’s a great social experiment. At the time, I was wary of being like, “This is the end of Twitter.” And you’re starting to see now just how resilient it is.
We’re so caught up in it as reporters. We’re on it all the time. But there are normal people out there that still go to X because their soccer community is on it. Or it’s where they talk about movies and they have their six best friends there. It’s still a very sticky platform, in spite of the CSAM from Grok or the abusive stuff from Elon or the hate speech. They persist because it’s just where they’ve learned to be. It’s hard to generate that from scratch, and I think that’s why you’re seeing things like Bluesky hit a ceiling here in terms of attracting a wider audience.
Do you think as we go into the IPO and SpaceX becomes a public company and X is just one piece of the puzzle, that it increases or decreases in importance to Elon? It is his favorite thing. But running a public rocket company is just a very different set of priorities.
Maybe, but also he never operates as we expect him to. Running Tesla and having a Twitter account have not worked out well for him in the past. He’s been sued for some of the stuff he’s put on Twitter. I think of 2018 when he got sued for saying funding was secured for taking Tesla private. And he paid $20 million in a fine and largely got away with it.
He’s way more powerful now and way richer than ever. So I don’t think SpaceX being a public company will necessarily change his habits on X these days. The other day I saw him posting about the Anthropic deal. I don’t know if you saw that. That was in the S-1 and he was pushing back on the idea that Anthropic would pay this amount of money for a certain amount of years — this large amount of revenue, about $1.25 billion a year.
He was openly contradicting what was in the company’s IPO documents, which you cannot do during a quiet period. I don’t think anything’s going to come of that. I don’t think he’ll get a slap on the wrist or anything. He’s just higher than any form of accountability right now.
That brings us to the larger SpaceX IPO in general, because the whole thing is structured to avoid the mechanisms of accountability that usually exist in our markets. It’s going to end up on the NASDAQ in some way, shape or form in a way that basically all of us are going to end up invested in SpaceX.
And we can’t take our dollars away because it’ll be in index funds. Elon is going to control an enormous part of the company in a way that maybe he just can never be removed. Who knows if even having a board of directors is important in that case. And then he has a monopoly on rocket launches, at least for now, and who knows if there will be market alternatives that provide accountability to SpaceX.
Walk us through how this is structured. You’ve written about it at length that the SpaceX IPO is a corporate governance disaster, if you care about corporate governance. Walk us through it.
I’m a big corporate governance guy myself.
[Laughs] It is the hottest. It’s what all the TikTok dances are about lately.
Yeah, I have a corporate governance tattoo on my lower back. But no, this is serious stuff and it’s concerning to people that study corporate governance. So let’s talk about Elon Musk’s ownership of the company and his voting control of the company.
He has super-voting shares that, all told, give him about 85 percent control of votes at the company. And that’s a super-supermajority at this point. He basically controls every corporate decision at the share voting level. I think of something like Meta, for example, and compare that to Mark Zuckerberg. With super voting and voting agreements, Zuckerberg controls about 60 percent. Elon has even a larger stranglehold on his company than that. What does that mean? It means he controls the board, he appoints friends and advisors to board seats.
There’s no independent board commission to structure pay packages. Essentially he has control over how he gets compensated. I wrote about this pay package that he got earlier this year where he was awarded 1.3 billion shares in what’s called restricted stock.
If you actually look at the footnotes there in that S-1, you could see that he’s already able to vote that stock, which is insane and it’s unheard of. He hasn’t earned any of these shares and these shares are pegged to hitting milestones with the company, like creating a colony on Mars with a million people and putting data centers in space with, I think, 100 terawatts of compute a year, just an astronomical figure.
He has to hit these things in order to gain these shares to sell them. Well, he hasn’t hit any of these milestones and he’s able to vote these already given the stipulations that were put on them from management.
Can I ask you about this, the colony on Mars? This is all bananas, right?
I guess we’re burying the lede here, yeah.
He gets a huge pay package if he puts a colony on Mars with a million people in it and puts however many terawatts of compute in space. He’s in charge of this S-1. He obviously wrote it to his own specifications. Why set milestones that are unachievable and then vote the stock anyway instead of just giving yourself the stock?
Good question. Maybe it gives it some semblance that he has to work towards these goals, but if you talk to corporate governance folks, they’re appalled that he gets to vote these anyways. This adds to his voting control, that 85 percent we talked about earlier. On top of that, he gets to take out loans against these shares. Of course that comes with board approval, but he controls the board. So he’s able to take out loans against these shares and get cash. And yeah, I don’t know. I don’t know why he’s even playing this dance of “I have to hit these milestones.”
My theory is that it’s so he can tweet about them. Legitimately, my theory is that he wants to be able to say, “I don’t get paid unless I put a million people on Mars.” Regardless of the technical details of whether he can vote the shares or take out loans against them as collateral. He gets to represent to the world, “I don’t receive the windfall until there are a million people on Mars.”
That’s a great point. There’s also a fun little wrinkle here in that you don’t pay taxes on them until you earn them. And so because he hasn’t earned them, because he doesn’t technically hold them yet or doesn’t have the ability to sell them, it’s not taxable. He doesn’t have to pay taxes on that either until he hits his milestones. In some ways, if you believe he’ll never hit them, he can still derive the power and potentially financial gain from it simply by holding them, or having this pay package under his thumb.
The usual way that a market could correct this is by a bunch of people telling retail investors, “Don’t invest in this IPO.” Or a bunch of people who are angry at Elon Musk selling their Teslas and telling their friends not to buy this IPO and driving their stock price down. That is a market corrective. We can see this happen with a variety of companies over time.
This IPO is going to end up in index funds very quickly in a way that I think is also terrifying a bunch of corporate governance experts, the sexiest new characters in American politics. Talk about that. How are we going to mechanically all end up owning SpaceX whether we want to or not?
This is the most underplayed and insane thing to think about. So how does this work? Maybe you and I take a look at the SpaceX IPO and we say, “I don’t like those financials. I personally would not invest in that company. I’m not going to buy shares through my Robinhood account or Schwab account.”
But let’s say we also have investments in index funds. We all have retirement accounts, and sometimes they’re invested in index funds. So we’re just passive investors in stock and these index funds trace the American industry by buying up shares in various stocks. So what’s happening here is that SpaceX and these indices have worked together to relax some of the rules. Let’s take the NASDAQ-100, for example. The NASDAQ-100 has 100 stocks in it that trace American industry, like blue chip stocks.
Typically, it takes about 90 days for a company after its IPO to enter the NASDAQ-100, or to be allowed into the NASDAQ-100. The reason for that is it allows for some type of cooling period. Typically, after an IPO’s stocks go up and down, it takes some time to settle on the public market and to have a steady valuation, and that’s normal.
In this case, SpaceX gets to enter the index after 15 days, in the midst of this very big hype cycle. That’s essentially going to force a lot of these index funds to buy up SpaceX shares because it has entered the index. That will give SpaceX access to a lot of capital it would have had to have waited for for a couple of months. And that’s going to continue to drive the buying frenzy in the stock. It’s genius in a way, from SpaceX’s perspective, to get that access to billions of dollars of capital in some of these index funds, and that’s just one way the rules have been relaxed for this IPO.
How did that occur? Did Elon just roll up to all the index fund owners and say, “Pretty please?” Did you buy them off? Was it above board? Was it corrupt?
Basically it’s been pretty opaque, but these things were simply announced. There have been rules around profitability that have been relaxed as well at some of these indices. There have been rules around governance that have been relaxed as well.
What I’m seeing here, what I’ve talked to folks about, is just this level of hype has generated so much FOMO around this stock, this fear of missing out. You get to the point where these rules are being thrown out. I talked to one corporate governance expert who said it’d be like having all the rules and setting aside all the rules for football. And you play with the rules of football and you’ve perfected it and you have it down to the rules of the game and then when you get to the Super Bowl, the biggest event of the year, you change those rules. And that’s what we’re seeing here with SpaceX.
There are a lot of people who root against the Kansas City Chiefs who understand exactly what you’re talking about in very specific ways.
[Laughs] It’d be like taking the tush push out of the NFL.
Well, I get it, but it’s also one of the biggest IPOs. The banks have all been listed on the IPO. They’re all participating in it. Is it just as simple as they all want the business, or do they also have some amount of influence over the index funds so that they’re just changing the rules?
Yeah. I think of this quote from a story that two of my colleagues wrote last week for The Times. There’s this one fund manager who simply said, “If I miss out on the SpaceX IPO, someone’s going to tap me on the shoulder and ask me why I wasn’t in that. Whereas if I get burned on the SpaceX IPO, so many other people are going to get burned as well. So I have a way to cover my ass.”
You’re just reading this quote and thinking, “What is going on here? He’s just admitting to a herd mentality here.” That’s what we’re seeing. Now multiply that by every retail investor who’s getting marketing materials on Robinhood telling him, “Oh, we have IPO shares available in SpaceX, buy, buy, buy.” It’s unheard of.
One of the things that really strikes me about that is the normal market dynamic is some people would obviously heavily bet on SpaceX succeeding and some people would heavily bet against it and you want that dynamic to find the right price for this. Here that just seems to be erased.
I’ll be interested in the short interest against this stock. I think that’ll be very interesting. But if you look at Elon’s track record, let’s say with Tesla for example, and how the stock has gone up over the years there, he’s completely crushed a lot of shorts there. He used to go to war against them, and he used to tweet about them all the time, but the best way to beat short sellers is to continue to increase the value of the stock, which Tesla has done over the years.
There’s just a meme-ification of this whole thing. This is not just a hype stock, but a meme stock in some ways, and that’s what happens when you have a celebrity CEO like this running a company.
That’s what’s really interesting to me because the car sales are falling. The product is not as successful as it once was. In many cases, the products are old. The Model S and Model X are being discontinued because they’re so old and he doesn’t want to spend money updating them.
Now, he’s promising robotics and robotaxis and a bunch of other things that may never come to pass. Is he going to be able to pull the same move with SpaceX? Just continually promise something bigger to come in the future that changes the value dynamic with the company?
He is right now as we speak. What happened at the beginning of this year? SpaceX was going along its way. It was a launch business with rockets that have self-landing capabilities and a really good business in Starlink. And what did he do? He combined it with xAI and said, “Actually, you know what we’re going to do? We’re going to put data centers into space and this is the future. And oh, by the way, we’re going to put a factory on Mars to build these satellites to launch into space. And then we’ll get to the Mars colony.”
These are goals that have come up within the last year. He didn’t talk about these things previously. In the same way at Tesla where he has completely pivoted the company towards robots and the humanoid whatever things, you’re getting the same effect at SpaceX where he’s just selling people on a completely different bill of goods.
It’s just so interesting. I look at some of the contradictions he’s made over the years. There’s a tweet of his that he put up probably a year ago where he said, “The Moon doesn’t matter, we’re not focused on the Moon, we’re focused on Mars.” And then you go back and you look at the IPO documents and what he said more recently in the last couple of months and now they’re all in on the Moon. And that’s because NASA has put a renewed focus on the Moon and there’s money there.
So yeah, if you’re going off of what Elon says, it is whichever way the wind blows at this point and thus far that’s worked for him. People are willing to go with him and believe in him.
Here we are, a half hour into a conversation about the SpaceX IPO and we’re going to talk about the fundamentals of the SpaceX business because that’s about where it ranks. It’s like the 15th thing on the priority list when you talk about the SpaceX IPO is the fundamentals of the business.
As you’ve said several times now, Starlink is the only profitable part of this business. It generated $11.4 billion in revenue last year. It goes up and down. Everything else is a gigantic money loser. The AI division had a deficit of $6.4 billion. The NASA contracts for launch lost $657 million.
Everything else is losing money and then Starlink is the business that’s growing and generating actual profits. I look at that, I think, “Boy, I’ve covered the broadband industry for a long time here at The Verge. AT&T and Verizon are not the world’s sexiest businesses. They’re not throwing off so much margin that you can lose $6 billion on AI for the rest of your life.”
How does that work? Is there more Starlink to be had? Are we going to rip up all the fiber in the world and we’ll all get satellites? How do you generate enough money with Starlink to pay for all of this other stuff?
Great question. I believe SpaceX thinks Starlink can continue to grow. There are a lot of markets that haven’t been tapped yet. I think of something like India, for example, where the company is heavily courting the Modi government there to allow them to operate in a country with 1.5 billion people. There are markets like that where it can access and continue to grow that roughly 10 million, I think, monthly active user base.
Can I push back on that just for one second?
Sure.
The Indian market is very complicated, but it is very well served by its own telecom providers. Reliance Jio is the winner in the Indian market and a huge number of people just have a cell phone as their primary connectivity device. They’re doing fine and it’s dirt cheap. Even if you’re excited about putting Starlink in that market, how do you compete against that? Is it possible? Have they laid out the case?
They have not. You can also argue that the revenue per user there is not going to be the same as it would be in the US or wherever else. But yeah, they’ve made the argument that as long as it continues to grow, it’s a good thing. They’ll continue to launch more satellites into space with these things and cover the world essentially.
SpaceX CEO Gwynne Shotwell gave a presentation at Mobile World Congress earlier this year where she basically put out a hit on all these big telecoms. Online she’s compared herself or compared SpaceX to David versus Goliath, which is a convenient narrative where you have a $1.25 trillion David going against these supposed Goliaths here.
That’s the bull case for Starlink. But if you look at the other fundamentals of this, the spending on AI is quite nuts. It’s losing so much money on AI development. We haven’t even talked about the massive amount it has to pay for Cursor, which is a $60 billion deal. So you asked earlier, what are people investing in here? They’re investing in promises. There are no fundamentals here.
We can talk until we’re blue in the face about profits and revenues and growth, but at the end of the day, most investors are betting on Elon’s words and his ability to sell them on this idea of putting data centers into space or getting people to Mars.
The AI piece is fascinating. They’re estimating that $22.7 trillion will be generated from enterprise AI.
Yeah, I want to talk about this TAM. It’s just insane.
TAM stands for total addressable market. It is $28 trillion, I think, maybe slightly more.
There’s a great line in the S-1, which says something like this is the largest TAM ever in human history. And I’m like, “Cool. Show me how you got there.” And it’s just like, “Trust me, bro.” It’s like, “We got this. We did the numbers and it’s $23 trillion in AI and $3 trillion in rocket launches.” I don’t know where the fundamentals are for that. What are they basing that on? Sure, it’s in their S-1, but it’s a lot of “trust me” at this point.
The case for this is going to be a great IPO because SpaceX figured out the Falcon 9 and rocket reusability and they essentially have a monopoly on launched services in the United States, at least until Jeff Bezos and Blue Origin figure out whatever they’re going to figure out. That’s a pretty good case. I can see that case.
Starlink is a growing business. We are essentially the default government contractor for a very important mission both in national security and telecom and everything else we use satellites for. I see all of that.
Why add on this totally illusory enterprise $22.7 trillion from AI when you’re up against OpenAI, Anthropic and Google? And Elon has admitted very publicly that xAI was not built correctly and needs to be totally rebuilt. Now, he’s selling compute capacity to Anthropic on the side.
TAM says everything. It shows where they think the addressable market is, which is in AI. And all the hype right now is around AI, with OpenAI and Anthropic both expected to go public or at least file to go public. Anthropic did that today. I think Elon saw that. And if you had just taken the old SpaceX business public, what does that look like? That’s the launch business and that’s Starlink, that’s a solid business, right?
Is it a $1.25 to $1.5 trillion business? No. But if you layer on all these promises of, “Actually, we control getting things into space so we’re going to control getting data centers into space, and we’re going to own all the data centers in space and everyone’s going to have to rely on us to power the future of the American economy,” that’s a much more bullish proposition. It’s also a much more valuable proposition and one that you can raise a lot more money on.
Elon sees this as a singular opportunity to raise money. We’re talking about $50 to $75 billion in cash raised. That outstrips the current largest IPO, which was Saudi Aramco that raised almost $30 billion a couple of years ago. If you layer AI on top of that, you get a lot of this hype that he can sell into and raise all that cash.
What does he want to do with that cash?
Of course, get people to Mars.
Obviously.
[Laughs] Of course, easy. Once you have $50 billion, you can land a million people on Mars. Yeah, these are expensive propositions. Building Starship is an expensive proposition. Building data centers on Earth is an expensive proposition, as is buying up all that compute, buying up talent, and buying companies like Cursor. He needs that cash to do all those things and continue building up Musk Inc. It’s not like they’re going to sit on this cash pile for a long time.
It occurs to me that Elon does not enjoy normal rich guy activities. He’s not buying boats. He doesn’t have a fleet of cars. He does have a lot of children. They’re expensive. I’ve got two. They seem very expensive. I can use $50 billion just for that. The AI piece of it is confusing to me because it doesn’t seem like there’s evidence that Elon’s AI efforts are competitive.
They’re competitive in the fact that Grok has distribution on X and I think a lot of these companies would love to have distribution that way. The fact that X users can just talk to Grok whenever they want — to do some unsavory things, but they still have distribution. Claude does not have that distribution. But Anthropic is ahead in coding.
Anthropic is ahead in a lot of places. OpenAI has mindshare. Google has distribution in every way it can possibly have distribution. It’s going to power the next version of Siri for Apple. How do you win? How do you raise all this money and say, “We’re going to have a $22 trillion enterprise service at market,” when no one in AI thinks you’re even close to the lead right now. Can you just buy the talent?
xAI is clearly behind in the model war. I saw this great tweet the other day that if Claude is Coca-Cola and OpenAI is Pepsi, then Grok is the RC Cola.
I knew you were going to say RC Cola and I was pre-offended.
[Laughs] Oh, no. Are you an RC Cola fan? I could go with Faygo for the Juggalos out there and I’m sure there’s a lot of Juggalos that like using Grok.
But it’s the fourth- or fifth- or sixth-best model, at least in terms of popularity. Can you build a business around that? Can that business be worth trillions of dollars? Probably not. And there’s some admission from Elon that, or there has been admission from Elon that it hasn’t gone as well.
You mentioned how he said the company wasn’t built right. You look at the Cursor acquisition they’re trying to pay for to get back in the game. You also look at this very interesting deal that the company has with Anthropic to rent out its compute from one of its main data centers that it built in Tennessee.
Colossus has two data centers, Colossus 1 and Colossus 2. It has since rented out one of these to Anthropic in a $1.25 billion a month deal. Anthropic is paying that much to get access to that compute. You’d probably argue that if everything was going swell at Grok and at xAI that they’d be using all that compute to push their own models and help their own customers. But in this case, it’s become sort of an AWS-type service where it’s renting out its space and that’s a good business. I’m not going to deny that — who wouldn’t want $1.25 billion a month in revenue? But it’s not what that thing was built for in the first place.
Is there a way back for them to lead at the frontier? Is it, “We’re going to raise $50 billion and maybe we’ll just hire everybody from OpenAI”?
Potentially. And if you look at some of the comments last week from Elon on X where he pushed back on the idea that that Anthropic deal would be for the next three years, he said, “We reserve the right to take some of that compute back.”
He’s suggesting maybe their models get good again to the point where they’ll need that compute. He changes with the wind and his business plan changes with the wind. We’ve seen that in the last six months to a year, these completely new businesses are coming out of nowhere. I don’t think there’s ever a “never” or a “never again” for Elon, and I guess he reserves the right to return to that at some point.
I want to end where we started, which is in 2022 when Elon buys Twitter and renames it X. It’s now famously the “everything app.” We’re clearly all doing our payments there all day long. You wrote a whole book about it. I made the prediction that buying Twitter would trash his reputation and maybe harm his companies.
Here we are now on the cusp of what might be one of the biggest IPOs in history. And it seems like in order to make it work, all of the rules of the game have had to be changed or rigged to favor Elon. If we were operating in a normal circumstance with the normal rules, with the normal index fund seizing rules, and he had to wait 90 days for profitability and people are looking at the actual fundamentals of this business, do you think there’s a chance that this IPO is as big as it’s going to end up being?
Probably not as big. It’s hard to say. I still think there would be an incredible amount of hype around this company. You just don’t get this type of excitement for any CEO beyond Elon Musk. For a lot of people, they don’t necessarily pay attention to his politics or his everyday posting on Twitter, his hate speech or whatever thing he’s concocting on the platform. They see him as a successful businessman, a generational talent that put Teslas on the roads, and they see that every day.
They see that as a chance to invest in him. There would still be a large amount of retail, obviously not the same amount as having to force index funds to buy into a company. But that alone might drive a lot of success for this IPO. Again, it’s a completely hypothetical situation. We’ll have to see in two weeks.
I’m very curious. Do you think there are any other correctives? We’ve talked about the market correctives, you talked about the index rules, you talked about the corporate governance issues. Are there any other correctives here, or are we just along for the ride?
Oh man, I’ve thought about accountability for Elon for a long time. That’s the point of our book. How do you hold someone that rich accountable? And I just think the normal levers of accountability for someone like that have gone out the window. Yeah, we’re along for the ride.
I have one example in this IPO, which is if you’re a shareholder in SpaceX, you agree to arbitration for any issues around if you believe some kind of fraud or violation of securities law has happened. In the past, Elon has faced lawsuits from shareholders at Tesla and Twitter. At SpaceX, he’s essentially removed that ability to pursue those types of shareholder lawsuits.
He’s stacking the deck for himself here and removing any of the obstacles he could face as a public company CEO and entrenched himself in this company, built a pretty big moat around himself. The impact of that will be seen for years to come.
Well, Ryan, it feels like no matter what, you and I are both going to end up as SpaceX shareholders. So I’ll see you at the next meeting. Thank you so much for being on Decoder. This is great.
Thanks for having me. I’ll see you on Mars.
[Laughs] One million strong, bro.
Questions or comments? Hit us up at decoder@theverge.com. We really do read every email!
I was very confident that Musk would come to regret buying Twitter back then. I wrote a piece called “Welcome To hell, Elon,” which is probably the single most-read thing I’ve ever written. My thesis was that there would be no way to grow Twitter users and revenue without moderating the platform well and, ultimately, that Elon buying Twitter would destroy his reputation and cause damage to his other companies.
Now, we have the numbers from the SpaceX IPO filing to see how right my prediction was. X is shrinking by every major metric, but it may not matter, as Ryan points out. Take a listen, and let me know what you think.
Ryan and I also got into all those rules being broken to land the SpaceX IPO — rules about shareholder control, inclusion in the major index funds, and all the other levers of market accountability that usually serve to keep companies in check. You’re going to hear us say “corporate governance” a lot in this episode, and while it may sound boring, it won’t be if you take a shot every time it comes up.
Okay, don’t do that. But do consider what it means that Elon has become so rich, so powerful, and so detached from the levers of accountability that he can apparently get away with anything. That’s all without any of the major fund managers or investors calling foul because they don’t want to miss out on what could be the biggest financial windfall in recent memory. There’s a lot to think about in this episode.
Okay: New York Times tech reporter Ryan Mac, on Elon Musk, X, and the SpaceX IPO. Here we go.
This interview has been lightly edited for length and clarity.
Ryan Mac, you’re a technology reporter at the New York Times. Welcome to Decoder.
Thanks for having me.
I am really excited to talk to you. I can’t believe you’ve never been on the show before. I feel like we’ve done a lot of reporting in and around each other. I’m a big fan. Thanks so much for being on.
I know. What the hell, man? You just have avoided me this whole time. But no, I’m kidding. It’s good to be here. I’ve listened to many episodes, so great to be a part of it.
Well, now we’re going to ask you to answer for your crimes, which is what the Decoder audience really wants me to do, I guess.
Speaking of crimes, we’re going to talk about the SpaceX IPO. Elon Musk has obviously filed to take SpaceX public. There’s a lot in that IPO, including the idea that there’s a $28 trillion addressable market for SpaceX services, which is more than the world. Just a lot in there.
You’ve reported on a lot of it. So I do want to dive into it, but I actually want to start with X, the everything app, the app formerly known as Twitter. Because the SpaceX S-1 really gives us our first look into what that business is, what it has become, where it’s growing.
In 2022, I wrote an article — maybe the most viral article I’ve ever written — it was called “Welcome to hell, Elon,” in which I very confidently predicted that buying Twitter would be a disaster for Elon Musk. I’m just going to read you my thesis. It was the first sentence of the piece. And then I want to try to back into what we know about X. I’m very curious if you think this has come true or not.
So my thesis was: “Twitter is a disaster clown car company that is successful despite itself and there is no possible way to grow users and revenue without making a series of enormous compromises that will ultimately destroy Elon Musk’s reputation and possibly cause grievous damage to his other companies.”
There’s one view to say, “Yep, that totally came true.” There’s another view to say that actually Elon is more powerful than ever and on the cusp of an IPO that’s going to make him a trillionaire. So tell me about X. What do we know about X, the company in the years since Elon has bought it and what do we know about its financials as reported in this S-1?
Sure. I think you mentioned the word “growing” in all that and I think the place to start is the fact that X is simply not growing. It’s stagnated in terms of revenue, stagnated in terms of user growth. It’s been buried twice within Elon’s companies — first into xAI and now into SpaceX. So it’s become, in some ways, an afterthought in the Musk empire, despite it still being arguably Musk’s favorite thing. He spends countless hours a day on that thing, like many of us used to, and many of us still do.
But in terms of a business proposition, it’s a non-factor if you compare it to some of the other aspects of this business — something like Starlink, for example. If you look back at 2022, it’s just bizarre. He bought this company on a whim. He pitched this idea to investors that he would have one billion users. He would have integrated payments. It would be somewhere you could potentially book a taxi. He pitched this idea of it being WeChat. You mentioned the everything app. And it’s certainly not the everything app. At one point he was like, “You could watch TV on it.”
None of that has come to fruition. Yet I look at what’s happened in the last four to five years since then, and he’s gotten more powerful than ever. His net worth has increased. I think around the time he bought the company he was around $300 billion. His net worth now fluctuates anywhere from $600 to $800 billion these days. And a SpaceX IPO will take him potentially beyond the trillion dollar mark for the first time ever in human history. So it’s bizarre in that there are a lot of contradictory things about it, but at the end of the day, I’d argue he still comes out on top.
Is it just as simple as he bought a distribution platform for his own tweets and he controlled it and he fixed the algorithm to favor himself and that worked? And it doesn’t matter that revenue is down $100 million year over year and not even quite 40 percent of Twitter’s pre-acquisition revenue?
He’s destroyed the business by every metric we can see in the S-1. Every number is down. And only the revenue from data licensing to AI companies is up.
His own AI company too. Yes, if you singularly look at X as a business, it’s clearly a failure from the time he took over the company to Fidelity marking the valuation of the company down to $10 billion before he merged it with xAI. But also you have to look at it in the whole landscape of Musk Inc. Since he bought the company, he spun up xAI, raised billions of dollars for that company. He then merged it into xAI, burying it, and then he merged it again with xAI into SpaceX.
I guess he’s up, if you’re doing a plus-minus analysis of valuations of these companies. Again, these are valuations that seemingly have no basis in business fundamentals. We’re playing with Musk math here. He has this whole cadre of investors and friends that are willing to back him to the end of the Earth, but yeah, he’s, I’d say, winning.
There’s a version of this where you could straightforwardly make the argument that however many billions he lost on Twitter is worth it as an investment that got him to, “We’re going to do a trillion-dollar SpaceX IPO.”
Yeah, and I caution against looking back at it historically and thinking that was his plan all along. There is this trope that came up after Trump won the election that Elon Musk bought X to then help elect Donald Trump. There was really no proof of that, especially when we did our reporting at The Times and for our book Character Limit, but it has worked out for him. It’s indisputable in that sense.
He has bought a distribution platform for his own tweets. He’s the most followed person on the platform now. He controls the algorithm, he controls the content that gets boosted on the platform. I don’t know how to say this more, but he’s winning and that’s just where we are right now in society.
In the pre-X days when it was still Twitter, Elon would constantly talk about how he didn’t do any marketing for Tesla. They had spent no money in advertising, no money in marketing. He would just tweet to move Tesla sales up and down. And there was infinite demand for the Tesla Model 3 in particular and the Model Y. So much so that every other car maker essentially got confused and made Tesla Model Ys of their own.
They just thought people wanted electric cars. And maybe they just wanted Tesla’s and maybe there was a meme stock component to it, but Elon was just very good at using Twitter to drive demand for his products. He would constantly say, “I don’t have to pay for marketing. I just have this platform.”
The other part of my thesis was that buying X and changing the algorithm and being as political as he has been would cause reputational damage to Elon, would cause reputational damage to his companies. There’s some evidence that that is true in the case of Tesla, where the cars are less popular than they once were, certainly. There are campaigns to protest at Tesla dealerships. Is that true for SpaceX? Is it true for the rest of his empire?
It’s a great question. And with Tesla, of course, after the election, you saw the Tesla Takedown protests, people slapping those bumper stickers on their cars: “I bought this before Elon went crazy.” You could see the share of the Tesla market in the EV market falling. Of course, there’s a lot to do with the rise of Chinese vehicle manufacturers like BYD.
But I don’t know. I don’t think I see the same kind of reputational harm to SpaceX that I’ve seen with Tesla. And that may be because there’s just not as much consumer contact with parts of SpaceX. I think of something like the launch business. I’m not going out and buying a launch. Regular people aren’t buying launches; SpaceX is contracting with governments and big companies. Not to mention that they largely have a monopoly on getting things into space.
If your option is working with a company that has a CEO that is reputationally compromised versus not getting into space at all, you’ll probably go with the former. Starlink is another basket. It offers a product that is quite good and is not challenged in any way. I think of something like Iridium, for example, or the companies it competes against, and it just blows them out of the water. It’s such a strong service, so much so that governments rely on it and Ukraine relies on it and the numbers there continue to grow. That’s the crown jewel of the SpaceX business empire right now in terms of revenue and profit.
There’s no moral case to be made for being on Hughesnet still. You can’t be a better person by being on some of the other satellite providers that service the rural parts of the country. We see it in our own traffic and our own comments. We cover Starlink. There’s nothing better. They continue to innovate in the ways they continue to innovate and the audience doesn’t like it, but then there’s a huge part of the audience that says, “Wait, I don’t have a market alternative to this.”
You’re saying where there isn’t a market alternative, the reputational issues have not been a problem and every place where there might be or there’s a consumer market, the reputational issues have damaged it. I think X is actually the example of this. There are a number of market alternatives to X, so people have just left the platform.
Whether or not there’s one big competitor to X remains to be seen. Threads by some accounts is vastly bigger than X, but it does not have the influence. Bluesky is run by very charming, very ideological people. They’re doing whatever it is they’re doing. It’s certainly not a competitor head up to X. Is there a reason that influence hasn’t recapitulated itself anywhere, or that the people still on X have stayed there?
Oh man, it’s something I think about a lot and I think the rumors of X’s demise at the time were greatly exaggerated in a way. These social platforms are so sticky. I remember one of the rounds of layoffs — I think it was the “A Fork in the Road” or one of those earlier rounds — and Twitter went down for a period of time, too. I remember being at dinner and everyone just writing eulogies for Twitter that night. And I’m like, “Guys, I don’t think this is it.”
This is a pretty resilient platform. It’s been developed over more than a decade. It’s been around for a while. It’s quite a resilient thing and it has a dedicated user base. That’s why we love it. It’s a great social experiment. At the time, I was wary of being like, “This is the end of Twitter.” And you’re starting to see now just how resilient it is.
We’re so caught up in it as reporters. We’re on it all the time. But there are normal people out there that still go to X because their soccer community is on it. Or it’s where they talk about movies and they have their six best friends there. It’s still a very sticky platform, in spite of the CSAM from Grok or the abusive stuff from Elon or the hate speech. They persist because it’s just where they’ve learned to be. It’s hard to generate that from scratch, and I think that’s why you’re seeing things like Bluesky hit a ceiling here in terms of attracting a wider audience.
Do you think as we go into the IPO and SpaceX becomes a public company and X is just one piece of the puzzle, that it increases or decreases in importance to Elon? It is his favorite thing. But running a public rocket company is just a very different set of priorities.
Maybe, but also he never operates as we expect him to. Running Tesla and having a Twitter account have not worked out well for him in the past. He’s been sued for some of the stuff he’s put on Twitter. I think of 2018 when he got sued for saying funding was secured for taking Tesla private. And he paid $20 million in a fine and largely got away with it.
He’s way more powerful now and way richer than ever. So I don’t think SpaceX being a public company will necessarily change his habits on X these days. The other day I saw him posting about the Anthropic deal. I don’t know if you saw that. That was in the S-1 and he was pushing back on the idea that Anthropic would pay this amount of money for a certain amount of years — this large amount of revenue, about $1.25 billion a year.
He was openly contradicting what was in the company’s IPO documents, which you cannot do during a quiet period. I don’t think anything’s going to come of that. I don’t think he’ll get a slap on the wrist or anything. He’s just higher than any form of accountability right now.
That brings us to the larger SpaceX IPO in general, because the whole thing is structured to avoid the mechanisms of accountability that usually exist in our markets. It’s going to end up on the NASDAQ in some way, shape or form in a way that basically all of us are going to end up invested in SpaceX.
And we can’t take our dollars away because it’ll be in index funds. Elon is going to control an enormous part of the company in a way that maybe he just can never be removed. Who knows if even having a board of directors is important in that case. And then he has a monopoly on rocket launches, at least for now, and who knows if there will be market alternatives that provide accountability to SpaceX.
Walk us through how this is structured. You’ve written about it at length that the SpaceX IPO is a corporate governance disaster, if you care about corporate governance. Walk us through it.
I’m a big corporate governance guy myself.
[Laughs] It is the hottest. It’s what all the TikTok dances are about lately.
Yeah, I have a corporate governance tattoo on my lower back. But no, this is serious stuff and it’s concerning to people that study corporate governance. So let’s talk about Elon Musk’s ownership of the company and his voting control of the company.
He has super-voting shares that, all told, give him about 85 percent control of votes at the company. And that’s a super-supermajority at this point. He basically controls every corporate decision at the share voting level. I think of something like Meta, for example, and compare that to Mark Zuckerberg. With super voting and voting agreements, Zuckerberg controls about 60 percent. Elon has even a larger stranglehold on his company than that. What does that mean? It means he controls the board, he appoints friends and advisors to board seats.
There’s no independent board commission to structure pay packages. Essentially he has control over how he gets compensated. I wrote about this pay package that he got earlier this year where he was awarded 1.3 billion shares in what’s called restricted stock.
If you actually look at the footnotes there in that S-1, you could see that he’s already able to vote that stock, which is insane and it’s unheard of. He hasn’t earned any of these shares and these shares are pegged to hitting milestones with the company, like creating a colony on Mars with a million people and putting data centers in space with, I think, 100 terawatts of compute a year, just an astronomical figure.
He has to hit these things in order to gain these shares to sell them. Well, he hasn’t hit any of these milestones and he’s able to vote these already given the stipulations that were put on them from management.
Can I ask you about this, the colony on Mars? This is all bananas, right?
I guess we’re burying the lede here, yeah.
He gets a huge pay package if he puts a colony on Mars with a million people in it and puts however many terawatts of compute in space. He’s in charge of this S-1. He obviously wrote it to his own specifications. Why set milestones that are unachievable and then vote the stock anyway instead of just giving yourself the stock?
Good question. Maybe it gives it some semblance that he has to work towards these goals, but if you talk to corporate governance folks, they’re appalled that he gets to vote these anyways. This adds to his voting control, that 85 percent we talked about earlier. On top of that, he gets to take out loans against these shares. Of course that comes with board approval, but he controls the board. So he’s able to take out loans against these shares and get cash. And yeah, I don’t know. I don’t know why he’s even playing this dance of “I have to hit these milestones.”
My theory is that it’s so he can tweet about them. Legitimately, my theory is that he wants to be able to say, “I don’t get paid unless I put a million people on Mars.” Regardless of the technical details of whether he can vote the shares or take out loans against them as collateral. He gets to represent to the world, “I don’t receive the windfall until there are a million people on Mars.”
That’s a great point. There’s also a fun little wrinkle here in that you don’t pay taxes on them until you earn them. And so because he hasn’t earned them, because he doesn’t technically hold them yet or doesn’t have the ability to sell them, it’s not taxable. He doesn’t have to pay taxes on that either until he hits his milestones. In some ways, if you believe he’ll never hit them, he can still derive the power and potentially financial gain from it simply by holding them, or having this pay package under his thumb.
The usual way that a market could correct this is by a bunch of people telling retail investors, “Don’t invest in this IPO.” Or a bunch of people who are angry at Elon Musk selling their Teslas and telling their friends not to buy this IPO and driving their stock price down. That is a market corrective. We can see this happen with a variety of companies over time.
This IPO is going to end up in index funds very quickly in a way that I think is also terrifying a bunch of corporate governance experts, the sexiest new characters in American politics. Talk about that. How are we going to mechanically all end up owning SpaceX whether we want to or not?
This is the most underplayed and insane thing to think about. So how does this work? Maybe you and I take a look at the SpaceX IPO and we say, “I don’t like those financials. I personally would not invest in that company. I’m not going to buy shares through my Robinhood account or Schwab account.”
But let’s say we also have investments in index funds. We all have retirement accounts, and sometimes they’re invested in index funds. So we’re just passive investors in stock and these index funds trace the American industry by buying up shares in various stocks. So what’s happening here is that SpaceX and these indices have worked together to relax some of the rules. Let’s take the NASDAQ-100, for example. The NASDAQ-100 has 100 stocks in it that trace American industry, like blue chip stocks.
Typically, it takes about 90 days for a company after its IPO to enter the NASDAQ-100, or to be allowed into the NASDAQ-100. The reason for that is it allows for some type of cooling period. Typically, after an IPO’s stocks go up and down, it takes some time to settle on the public market and to have a steady valuation, and that’s normal.
In this case, SpaceX gets to enter the index after 15 days, in the midst of this very big hype cycle. That’s essentially going to force a lot of these index funds to buy up SpaceX shares because it has entered the index. That will give SpaceX access to a lot of capital it would have had to have waited for for a couple of months. And that’s going to continue to drive the buying frenzy in the stock. It’s genius in a way, from SpaceX’s perspective, to get that access to billions of dollars of capital in some of these index funds, and that’s just one way the rules have been relaxed for this IPO.
How did that occur? Did Elon just roll up to all the index fund owners and say, “Pretty please?” Did you buy them off? Was it above board? Was it corrupt?
Basically it’s been pretty opaque, but these things were simply announced. There have been rules around profitability that have been relaxed as well at some of these indices. There have been rules around governance that have been relaxed as well.
What I’m seeing here, what I’ve talked to folks about, is just this level of hype has generated so much FOMO around this stock, this fear of missing out. You get to the point where these rules are being thrown out. I talked to one corporate governance expert who said it’d be like having all the rules and setting aside all the rules for football. And you play with the rules of football and you’ve perfected it and you have it down to the rules of the game and then when you get to the Super Bowl, the biggest event of the year, you change those rules. And that’s what we’re seeing here with SpaceX.
There are a lot of people who root against the Kansas City Chiefs who understand exactly what you’re talking about in very specific ways.
[Laughs] It’d be like taking the tush push out of the NFL.
Well, I get it, but it’s also one of the biggest IPOs. The banks have all been listed on the IPO. They’re all participating in it. Is it just as simple as they all want the business, or do they also have some amount of influence over the index funds so that they’re just changing the rules?
Yeah. I think of this quote from a story that two of my colleagues wrote last week for The Times. There’s this one fund manager who simply said, “If I miss out on the SpaceX IPO, someone’s going to tap me on the shoulder and ask me why I wasn’t in that. Whereas if I get burned on the SpaceX IPO, so many other people are going to get burned as well. So I have a way to cover my ass.”
You’re just reading this quote and thinking, “What is going on here? He’s just admitting to a herd mentality here.” That’s what we’re seeing. Now multiply that by every retail investor who’s getting marketing materials on Robinhood telling him, “Oh, we have IPO shares available in SpaceX, buy, buy, buy.” It’s unheard of.
One of the things that really strikes me about that is the normal market dynamic is some people would obviously heavily bet on SpaceX succeeding and some people would heavily bet against it and you want that dynamic to find the right price for this. Here that just seems to be erased.
I’ll be interested in the short interest against this stock. I think that’ll be very interesting. But if you look at Elon’s track record, let’s say with Tesla for example, and how the stock has gone up over the years there, he’s completely crushed a lot of shorts there. He used to go to war against them, and he used to tweet about them all the time, but the best way to beat short sellers is to continue to increase the value of the stock, which Tesla has done over the years.
There’s just a meme-ification of this whole thing. This is not just a hype stock, but a meme stock in some ways, and that’s what happens when you have a celebrity CEO like this running a company.
That’s what’s really interesting to me because the car sales are falling. The product is not as successful as it once was. In many cases, the products are old. The Model S and Model X are being discontinued because they’re so old and he doesn’t want to spend money updating them.
Now, he’s promising robotics and robotaxis and a bunch of other things that may never come to pass. Is he going to be able to pull the same move with SpaceX? Just continually promise something bigger to come in the future that changes the value dynamic with the company?
He is right now as we speak. What happened at the beginning of this year? SpaceX was going along its way. It was a launch business with rockets that have self-landing capabilities and a really good business in Starlink. And what did he do? He combined it with xAI and said, “Actually, you know what we’re going to do? We’re going to put data centers into space and this is the future. And oh, by the way, we’re going to put a factory on Mars to build these satellites to launch into space. And then we’ll get to the Mars colony.”
These are goals that have come up within the last year. He didn’t talk about these things previously. In the same way at Tesla where he has completely pivoted the company towards robots and the humanoid whatever things, you’re getting the same effect at SpaceX where he’s just selling people on a completely different bill of goods.
It’s just so interesting. I look at some of the contradictions he’s made over the years. There’s a tweet of his that he put up probably a year ago where he said, “The Moon doesn’t matter, we’re not focused on the Moon, we’re focused on Mars.” And then you go back and you look at the IPO documents and what he said more recently in the last couple of months and now they’re all in on the Moon. And that’s because NASA has put a renewed focus on the Moon and there’s money there.
So yeah, if you’re going off of what Elon says, it is whichever way the wind blows at this point and thus far that’s worked for him. People are willing to go with him and believe in him.
Here we are, a half hour into a conversation about the SpaceX IPO and we’re going to talk about the fundamentals of the SpaceX business because that’s about where it ranks. It’s like the 15th thing on the priority list when you talk about the SpaceX IPO is the fundamentals of the business.
As you’ve said several times now, Starlink is the only profitable part of this business. It generated $11.4 billion in revenue last year. It goes up and down. Everything else is a gigantic money loser. The AI division had a deficit of $6.4 billion. The NASA contracts for launch lost $657 million.
Everything else is losing money and then Starlink is the business that’s growing and generating actual profits. I look at that, I think, “Boy, I’ve covered the broadband industry for a long time here at The Verge. AT&T and Verizon are not the world’s sexiest businesses. They’re not throwing off so much margin that you can lose $6 billion on AI for the rest of your life.”
How does that work? Is there more Starlink to be had? Are we going to rip up all the fiber in the world and we’ll all get satellites? How do you generate enough money with Starlink to pay for all of this other stuff?
Great question. I believe SpaceX thinks Starlink can continue to grow. There are a lot of markets that haven’t been tapped yet. I think of something like India, for example, where the company is heavily courting the Modi government there to allow them to operate in a country with 1.5 billion people. There are markets like that where it can access and continue to grow that roughly 10 million, I think, monthly active user base.
Can I push back on that just for one second?
Sure.
The Indian market is very complicated, but it is very well served by its own telecom providers. Reliance Jio is the winner in the Indian market and a huge number of people just have a cell phone as their primary connectivity device. They’re doing fine and it’s dirt cheap. Even if you’re excited about putting Starlink in that market, how do you compete against that? Is it possible? Have they laid out the case?
They have not. You can also argue that the revenue per user there is not going to be the same as it would be in the US or wherever else. But yeah, they’ve made the argument that as long as it continues to grow, it’s a good thing. They’ll continue to launch more satellites into space with these things and cover the world essentially.
SpaceX CEO Gwynne Shotwell gave a presentation at Mobile World Congress earlier this year where she basically put out a hit on all these big telecoms. Online she’s compared herself or compared SpaceX to David versus Goliath, which is a convenient narrative where you have a $1.25 trillion David going against these supposed Goliaths here.
That’s the bull case for Starlink. But if you look at the other fundamentals of this, the spending on AI is quite nuts. It’s losing so much money on AI development. We haven’t even talked about the massive amount it has to pay for Cursor, which is a $60 billion deal. So you asked earlier, what are people investing in here? They’re investing in promises. There are no fundamentals here.
We can talk until we’re blue in the face about profits and revenues and growth, but at the end of the day, most investors are betting on Elon’s words and his ability to sell them on this idea of putting data centers into space or getting people to Mars.
The AI piece is fascinating. They’re estimating that $22.7 trillion will be generated from enterprise AI.
Yeah, I want to talk about this TAM. It’s just insane.
TAM stands for total addressable market. It is $28 trillion, I think, maybe slightly more.
There’s a great line in the S-1, which says something like this is the largest TAM ever in human history. And I’m like, “Cool. Show me how you got there.” And it’s just like, “Trust me, bro.” It’s like, “We got this. We did the numbers and it’s $23 trillion in AI and $3 trillion in rocket launches.” I don’t know where the fundamentals are for that. What are they basing that on? Sure, it’s in their S-1, but it’s a lot of “trust me” at this point.
The case for this is going to be a great IPO because SpaceX figured out the Falcon 9 and rocket reusability and they essentially have a monopoly on launched services in the United States, at least until Jeff Bezos and Blue Origin figure out whatever they’re going to figure out. That’s a pretty good case. I can see that case.
Starlink is a growing business. We are essentially the default government contractor for a very important mission both in national security and telecom and everything else we use satellites for. I see all of that.
Why add on this totally illusory enterprise $22.7 trillion from AI when you’re up against OpenAI, Anthropic and Google? And Elon has admitted very publicly that xAI was not built correctly and needs to be totally rebuilt. Now, he’s selling compute capacity to Anthropic on the side.
TAM says everything. It shows where they think the addressable market is, which is in AI. And all the hype right now is around AI, with OpenAI and Anthropic both expected to go public or at least file to go public. Anthropic did that today. I think Elon saw that. And if you had just taken the old SpaceX business public, what does that look like? That’s the launch business and that’s Starlink, that’s a solid business, right?
Is it a $1.25 to $1.5 trillion business? No. But if you layer on all these promises of, “Actually, we control getting things into space so we’re going to control getting data centers into space, and we’re going to own all the data centers in space and everyone’s going to have to rely on us to power the future of the American economy,” that’s a much more bullish proposition. It’s also a much more valuable proposition and one that you can raise a lot more money on.
Elon sees this as a singular opportunity to raise money. We’re talking about $50 to $75 billion in cash raised. That outstrips the current largest IPO, which was Saudi Aramco that raised almost $30 billion a couple of years ago. If you layer AI on top of that, you get a lot of this hype that he can sell into and raise all that cash.
What does he want to do with that cash?
Of course, get people to Mars.
Obviously.
[Laughs] Of course, easy. Once you have $50 billion, you can land a million people on Mars. Yeah, these are expensive propositions. Building Starship is an expensive proposition. Building data centers on Earth is an expensive proposition, as is buying up all that compute, buying up talent, and buying companies like Cursor. He needs that cash to do all those things and continue building up Musk Inc. It’s not like they’re going to sit on this cash pile for a long time.
It occurs to me that Elon does not enjoy normal rich guy activities. He’s not buying boats. He doesn’t have a fleet of cars. He does have a lot of children. They’re expensive. I’ve got two. They seem very expensive. I can use $50 billion just for that. The AI piece of it is confusing to me because it doesn’t seem like there’s evidence that Elon’s AI efforts are competitive.
They’re competitive in the fact that Grok has distribution on X and I think a lot of these companies would love to have distribution that way. The fact that X users can just talk to Grok whenever they want — to do some unsavory things, but they still have distribution. Claude does not have that distribution. But Anthropic is ahead in coding.
Anthropic is ahead in a lot of places. OpenAI has mindshare. Google has distribution in every way it can possibly have distribution. It’s going to power the next version of Siri for Apple. How do you win? How do you raise all this money and say, “We’re going to have a $22 trillion enterprise service at market,” when no one in AI thinks you’re even close to the lead right now. Can you just buy the talent?
xAI is clearly behind in the model war. I saw this great tweet the other day that if Claude is Coca-Cola and OpenAI is Pepsi, then Grok is the RC Cola.
I knew you were going to say RC Cola and I was pre-offended.
[Laughs] Oh, no. Are you an RC Cola fan? I could go with Faygo for the Juggalos out there and I’m sure there’s a lot of Juggalos that like using Grok.
But it’s the fourth- or fifth- or sixth-best model, at least in terms of popularity. Can you build a business around that? Can that business be worth trillions of dollars? Probably not. And there’s some admission from Elon that, or there has been admission from Elon that it hasn’t gone as well.
You mentioned how he said the company wasn’t built right. You look at the Cursor acquisition they’re trying to pay for to get back in the game. You also look at this very interesting deal that the company has with Anthropic to rent out its compute from one of its main data centers that it built in Tennessee.
Colossus has two data centers, Colossus 1 and Colossus 2. It has since rented out one of these to Anthropic in a $1.25 billion a month deal. Anthropic is paying that much to get access to that compute. You’d probably argue that if everything was going swell at Grok and at xAI that they’d be using all that compute to push their own models and help their own customers. But in this case, it’s become sort of an AWS-type service where it’s renting out its space and that’s a good business. I’m not going to deny that — who wouldn’t want $1.25 billion a month in revenue? But it’s not what that thing was built for in the first place.
Is there a way back for them to lead at the frontier? Is it, “We’re going to raise $50 billion and maybe we’ll just hire everybody from OpenAI”?
Potentially. And if you look at some of the comments last week from Elon on X where he pushed back on the idea that that Anthropic deal would be for the next three years, he said, “We reserve the right to take some of that compute back.”
He’s suggesting maybe their models get good again to the point where they’ll need that compute. He changes with the wind and his business plan changes with the wind. We’ve seen that in the last six months to a year, these completely new businesses are coming out of nowhere. I don’t think there’s ever a “never” or a “never again” for Elon, and I guess he reserves the right to return to that at some point.
I want to end where we started, which is in 2022 when Elon buys Twitter and renames it X. It’s now famously the “everything app.” We’re clearly all doing our payments there all day long. You wrote a whole book about it. I made the prediction that buying Twitter would trash his reputation and maybe harm his companies.
Here we are now on the cusp of what might be one of the biggest IPOs in history. And it seems like in order to make it work, all of the rules of the game have had to be changed or rigged to favor Elon. If we were operating in a normal circumstance with the normal rules, with the normal index fund seizing rules, and he had to wait 90 days for profitability and people are looking at the actual fundamentals of this business, do you think there’s a chance that this IPO is as big as it’s going to end up being?
Probably not as big. It’s hard to say. I still think there would be an incredible amount of hype around this company. You just don’t get this type of excitement for any CEO beyond Elon Musk. For a lot of people, they don’t necessarily pay attention to his politics or his everyday posting on Twitter, his hate speech or whatever thing he’s concocting on the platform. They see him as a successful businessman, a generational talent that put Teslas on the roads, and they see that every day.
They see that as a chance to invest in him. There would still be a large amount of retail, obviously not the same amount as having to force index funds to buy into a company. But that alone might drive a lot of success for this IPO. Again, it’s a completely hypothetical situation. We’ll have to see in two weeks.
I’m very curious. Do you think there are any other correctives? We’ve talked about the market correctives, you talked about the index rules, you talked about the corporate governance issues. Are there any other correctives here, or are we just along for the ride?
Oh man, I’ve thought about accountability for Elon for a long time. That’s the point of our book. How do you hold someone that rich accountable? And I just think the normal levers of accountability for someone like that have gone out the window. Yeah, we’re along for the ride.
I have one example in this IPO, which is if you’re a shareholder in SpaceX, you agree to arbitration for any issues around if you believe some kind of fraud or violation of securities law has happened. In the past, Elon has faced lawsuits from shareholders at Tesla and Twitter. At SpaceX, he’s essentially removed that ability to pursue those types of shareholder lawsuits.
He’s stacking the deck for himself here and removing any of the obstacles he could face as a public company CEO and entrenched himself in this company, built a pretty big moat around himself. The impact of that will be seen for years to come.
Well, Ryan, it feels like no matter what, you and I are both going to end up as SpaceX shareholders. So I’ll see you at the next meeting. Thank you so much for being on Decoder. This is great.
Thanks for having me. I’ll see you on Mars.
[Laughs] One million strong, bro.
Questions or comments? Hit us up at decoder@theverge.com. We really do read every email!