---
title: 'They''re getting Left Behind on SpaceX Stock'
source: 'https://youtube.com/watch?v=IkWPVpey5mA'
video_id: 'IkWPVpey5mA'
date: 2026-09-14
duration_sec: 997
channel: 'Meet Kevin'
---

# They're getting Left Behind on SpaceX Stock

> Source: [They're getting Left Behind on SpaceX Stock](https://youtube.com/watch?v=IkWPVpey5mA)

## Summary

This video analyzes SpaceX's valuation after its AI pivot, using Barron's mea culpa as a springboard. The creator breaks down how revenue and operating income estimates have exploded, and presents multiple valuation scenarios for 2027 and 2030.

### Key Points

- **Barron's Mea Culpa** [00:00] — Barron's admits they were wrong about SpaceX, citing the AI business transformation.
- **The AI Pivot** [02:10] — SpaceX pivoted from frontier AI spending to selling compute and building Grokbot, an agentic service.
- **Estimate Explosion** [04:28] — 2030 operating income estimates jumped from $4B (most bullish) to $197B consensus.
- **AI Infrastructure Contribution** [06:35] — AI infrastructure operating income is now projected at $178B by 2030.
- **Revenue Forecast Jump** [07:14] — 2031 AI revenue estimates rose from $150B to $530B, a 3.5x increase.
- **Margin Expansion** [07:45] — Implied operating margin at $197B is 47%, reflecting massive margin expansion.
- **Frontier vs. Application** [08:24] — The creator argues the application play (Grokbot, Palantir) is where the money is, not frontier labs.
- **Valuation Scenarios** [10:37] — At a 2.25 PEG and 30% growth, fair value in 2030 is $1,408; with Apple-level PEG, $1,500.
- **Current Valuation** [13:32] — SpaceX trades at 34x 2027 EBITDA, implying a PEG under one.

### Conclusion

SpaceX's valuation has been transformed by its AI pivot, with consensus estimates now supporting a fair value of $1,500 by 2030. Despite the hype, the stock may still be undervalued relative to its growth.

## Transcript

Parents just reported that they were wrong about SpaceX. They literally wrote here, Boy, we were wrong about SpaceX. And in this, we're going to go through what's actually going on with the valuation for SpaceX.
What's the valuation today? What's the fair price for it between now and the end of 2027? What's the potentially fair price for it by the end of 2030? And, as we mentioned last week,
how do those higher price targets fit in with what analysts' expectations are now for SpaceX revenues? And where are those coming from?
Where are those margin changes coming from? And how much are you actually paying for the space business versus the AI infrastructure business? We're going to break all of that down. But first, we're going to start with the Barron's piece here. The Barron's piece has the following to say.
they say that shortly before the SpaceX IPO we called SpaceX worth $90 per share now I gotta be clear I also talked about wanting to add to SpaceX
under $100 because I thought it was going down at least 30% post IPO and in fairness it did it ran from 135 down to 100
was a 30% decline but it ran up to 220 and then fell down to $104. But there was a rational basis for this, and you're going to see a similar argument here from Barron.
Keep in mind, just full disclosure, we still own a bunch of SpaceX in our VC fund, or otherwise. Full disclosure. At the time of the merger, Elon Musk's AI company was burning cash,
this is true, in an uncertain bid to cash AI leaders such as Anthropic. Correct. At the time of the IPO, it seemed like the goal was,
hey, we're going to blow money to try to have as advanced of a model as Anthropic or maybe a GPT Astra or Claude Fable or otherwise. That's a money-losing venture, in my opinion,
to try to catch up with these leading-edge models. What SpaceX did was they pivoted from, we're going to create the bleeding edge, So we're going to get into selling you an agentic service, Grokbot.
We'll talk about that in just a moment. I'm excited about Grokbot and also what Metamuse is doing. But while we do that, we'll also sell data center compute. And then they started inking contract after contract,
which has transformed the valuation trajectory of this company. You're going to see Barron's say exactly that here. Watch this. They say that now it looks like SpaceX shareholders are the winners of the XAI deal.
Well, it depends when they bought. If they bought after IPO, they're still upside down. You could basically buy lower today than what people were able to buy at IPO. Unless you've got an IPO allocation at 135, most people were able to start buying this at open trades at 155 to about 227,
which during that time post IPO you'd still be upside down today. I'll take a dollar count average down. But anyway, look at this. Barron says, what changed is the AI business.
Revenue has simply accelerated faster than anyone expected. That growth has radically altered near-term earnings forecasts and long-term cash flow projections. Yes, it is actually insane how much it's changed.
And you're going to see that. When we go to this spreadsheet that I'm going to show you, I'm going to show you exactly the data we used. This was the data that we used. I projected that by the end of 2030, we would see about $4 billion of operating income from launch services,
about $15 billion of operating income from Starlink. And because we thought they were going to blow money on frontier spending, we only saw AI infrastructure operating income of about billion The most bullish Wall Street estimate at that time was billion
So this was not an isolated, oh, be bearish on SpaceX. This was, everybody thought, okay, yeah, there's an AI infrastructure play here, but it's small.
Those estimates have moved from $4 billion on the most bullish to now consensus estimates of $197 billion.
At $2.5 billion and a peg ratio of about 2.25, and we'll play with these numbers in a moment, that brings you to a fair value in 2030 of $154.
If the stock today is priced at $150, you would be investing for a half percent return per year. That makes no sense. It makes no logical sense to buy it at $150 if that's your baseline assumption.
You would have to wait until about $68. And I'm not saying it goes to that. I'm saying that's where you would get a fair compensation for owning the stock. But these estimate changes have flipped this on its head.
You could see those estimate changes, just so you know where they're coming from, graphically on screen now. These are the estimate changes. We covered these last week, so I'm not going to rehash all of them.
I'm just going to go to the operating income estimate changes. The operating income changes for 2030 have gone from a low of about $40 billion at a low here around IPO time to 5x, $197 billion by 2030.
And before this, these estimates were even lower. which is the craziest part here. The estimates have exploded for operating income at this business.
And in fairness, this is the entire business, so we should back out the other portions of the business that we have here. So let's back out the other $19 billion. It's not going to make much of a difference.
So go $197 minus the $19 billion in the estimates from the other side. That brings you to about $197 minus $19. It brings me about $178 from AI operating infrastructure.
I'll show you what that does to the price targets in just a moment. But let's go back to the Barron's piece because they make some good arguments here. So we go to this Barron's piece. What's changed is that revenue has simply accelerated faster than anyone has expected.
Here we go. Around the time of the IPO, Wall Street forecasts for 2027 were basically operating income of $28 billion, or EBITDA, rather, of $28 billion.
Now for 2027, they've doubled. And by 2031, AI revenue alone, in early estimates, were around $150 billion.
By 2031 now, AI revenue estimates are $530 billion. So if you divide those into each other, that's about a 3.5x. So you can see how where the top line has exploded on a revenue expectation,
and the bottom line estimates have exploded, as we've seen in this transition of these charts right here. This is an explosion in operating income. If we look at the operating income margins, this is wild too.
Operating income estimates, 47% implied margin at $197 billion. is really, really remarkable, what the expectations have transitioned into.
And so Barron says, if you scroll down a little bit, some people aren't convinced to actually change their projections because they think that most of SpaceX's growth is coming from compute
rather than frontier AI. And they argue that that bearish for their AI story That their take I actually personally think that bullish for the SpaceX thesis
Here's why, and then we'll actually play with some numbers. I think there are two ways you can make money from artificial intelligence. I think there's the frontier play, and then there's the application play.
The application play is like your Metamuse, your Grokbot, your Palantir, and potentially other software plays, whether that's, you know, for now a Salesforce or whatever else, right?
These are your application plays. Grokbot is an application play. Frontier plays are things like Claude like Fable, right? So Claude Fable
or GPT Astra These are really expensive I don't want to invest in this because I think this is going to converge into a commodity
This is where the money is So Elon said, you know what? We'll go ahead and rent Compute to the Frontier and we will build the application layer. That is the most brilliant combination, in my opinion, and it is a shift
from the original estimates for SpaceX. Really remarkable. So, now that we understand how things have changed, how can we go look at a potential forecast for this business? Well, let's take a
look, and we'll show you pricing for today as well. So, if we change this segment here to 178, You know, that's your consensus. That's not even the bullish. That's your consensus.
So this could be even more crazy in a bullish scenario. Like, for example, we could ramp up Starlink income. We might do that in just a moment, just to see what the numbers would look like. But if I change this to $178 billion of operating income,
that gets us to our total operating income of $197 billion. That's the consensus estimate at the moment. at a 2.25 peg ratio, 30% forecast growth. That gets us in 2030 to $1,408.
But frankly, you could jack this up with this sort of EPS over here a lot. First of all, we could go to Apple level peg ratios of 2.24. Sorry, 2.4 I'll call it.
If I go to 2.4, that would be, we'll call it Apple level margins. around 27% bet. And if the growth at a terminal value was 30% in 2020-30 with these numbers,
you could argue $1,500. If the terminal growth rate is 50%, you could get to $2,500. It's insane. Now, another way you could analyze this, because that's spreadsheet math,
another way you could analyze this, and it's a little bit different with how you arrive at the numbers, but another way you could arrive at numbers like this is you could look at what EPS is projected to be. The EPS we have on this spreadsheet gets to a pretty rich and lofty almost $21 by 2030 on these bases.
If we use more of a Wall Street consensus, Wall Street forecasts right now are these EPS levels. If we take this growth rate right here, which averages out to about 85% over the next three years,
and instead we say, let's take 50% growth and start with 2027 earnings per share of just $1.74. At $1.74, we could get to a price target for 2027 of just $2.05.
That's not that juicy, right? $2.05, that's a fair price. So if anything you're buying under $2.05, fair price for 2027. Not personalized financial advice, obviously, right? But if these margins and growth rates can get maintained going into 2030 you can usually justify Or if you go all the way up to larger growth rates that So in other words
if you take a little bit of a blend of the Wall Street consensus estimates and forecast those growth rates out, $1,500 is easy by 2030. If you take the spreadsheet approach,
you can also get to $1,500 if we really think this compute infrastructure is going to print in this manner. And so I think a big example of
where you evaluate the AI data center place and where you evaluate if you want to put money into this is you look at a company like Oracle because at these numbers the company is actually trading for like a one peg
which is really cheap Which, ironically, is what Barron's admits in their article. If we go to, let me see where they mention valuation, valuation, valuation, valuation. They say, here we go.
What's more, the current valuation isn't that astronomical anymore. Now it's trading for just 34 times 2027 EBITDA. Right, and it's growing for more than that.
Or at a faster rate than that, significantly. So your peg ratio is under one. But as an investor, what you have to ask yourself is, does it make sense? Like, what do you want to invest in? Do you want to invest in the Frontier Labs?
Great. Then maybe Anthropix, your play, you know, whatever. Certainly NVIDIA, A&B, Broadcom, Marvell. Those are your data center plays now.
You could say even Bloom Energy is in there. if you want to forecast out to 2028 and you have balls of steel, you could argue N phases in there, you know, for solid state transformers. That's a high-risk call option.
If you want to play over here in the application layer, your play is software and, ironically, the compute holders.
So if you believe in this application layer, These could be the Microsofts, the Amazons, the Corweaves, the SpaceXes, to some extent the Dells for the software layer,
although they have some momentum risk with how much they've run lately. Of course, if you don't believe in either of these layers, then you don't want to touch any of those socks. You know, then you go buy real estate.
Which, hey, you know, I'm a big fan, obviously, of real estate. And that keeps you away from the forming bubble, so to speak.
But I still think it's early in this bubble. So, kind of incredible, but that gives you a little bit of a point of view of how SpaceX, ironically, looks cheap right now, and Barron's is coming out with their mea culpa.
And it makes sense because of those revisions, which are the same revisions that we've been talking about last week. Again, you look at, where was it, the SpaceX revisions.
I mean, you've seen them on screen a few times now, but I just want to go back and look at them one more time because there's some others as well. So if I go to the SpaceX revisions, it's not just the operating income.
Oh, yeah, yeah, it's the 2026 chart as well. Right, you look at the 2026 chart, and if you go into 2027, you've exploded in revenue estimates for 2017.
That's just getting started. Like 50 numbers have been published. You've had SpaceX come out and say, oh, we're signing even more deals. The CFO came out and said, we made a video on that CFO interview this weekend.
If you haven't seen it yet, you can watch it. During the advertising, these things that you told us here, I feel like nobody else knows about it. We'll try a little advertising and see how it goes. Congratulations, man. You've done so much. People love you. People look up to you.
Kevin Pass left their financial analyst. And YouTube.me, Kevin. Always great to get your take.