SpaceX just went public under the ticker SPCX.

We surveyed our RiskHedge members. About 80% of you said you won’t buy it today. Good choice.

As expected, SpaceX broke the record for the biggest IPO in history, raising $75 billion at a $1.77 trillion valuation.

But SpaceX quietly broke another record today, too. Among companies worth over $500 billion, SpaceX is now the most expensive in history—by a mile. Investors are paying nearly 100X sales for the stock.

What makes SpaceX worthy of such an eye-watering valuation?

Data centers in space.

• On March 21, Musk officially unveiled Terafab.

It’s a joint venture between SpaceX, Tesla (TSLA), and Intel Corp. (INTC) to build the largest chip-making plant in history.

Today, chipmaking is scattered across a complex global supply chain. One company designs the chip. Another prints it onto silicon wafers. Others handle testing... assembly... packaging and so on.

Terafab would consolidate every stage of chip production under one roof. That’s never been done before.

The goal is to produce 1 terawatt worth of artificial intelligence (AI) computing power per year. That's roughly 50X more AI horsepower than the world has today.

Why am I telling you this? Because Musk expects 80% of that output to go to space. SpaceX’s orbital data centers will do what AI computing hyperscalers like Amazon (AMZN) and Microsoft (MSFT) currently do on Earth.

SpaceX has already filed an application with the FCC for this...

• The plan? One million AI data center satellites in low-Earth orbit.

Similar to how Starlink satellites beam internet down to Earth, SpaceX’s orbital data centers would run AI workloads in space, then beam the data back to Earth.

And because they’re operating in orbit, they avoid many of Earth’s biggest problems...

A data center in orbit can tap directly into the largest power source available: the sun. It works 24/7/365. No land required. No NIMBY (“not in my backyard”) resistance. No competing with homes or factories for electricity.

The big issue keeping them back is launch costs.

Musk argues that he’ll bring those down enough so running AI compute in orbit will become cheaper than doing it on the ground within two to three years.

• There’s just one problem... Starship.

Everything rests on its success.

Starship is SpaceX’s massive next-gen rocket. It promises to make sending cargo to space 4X to 9X cheaper than it is today.

Starship is the missing ingredient that makes orbital AI data centers economical. It's also essential for building permanent bases on the Moon and eventually Mars… and for Starlink’s future.

The satellite internet business needs Starship’s enormous capacity if it wants to deploy next-generation satellites at scale. Starlink’s expansion slows otherwise. And its hyped direct-to-cell rollout becomes much harder to execute.

Bottom line: Without Starship, SpaceX isn’t a multitrillion-dollar space and AI company. It becomes just a rocket and broadband business. A great one, but an overvalued one.

• I’m extremely bullish on SpaceX long term.

I’ve met over a dozen current and former SpaceX engineers. Everyone agrees this is one of the great innovation stories of our lifetimes.

SpaceX still has huge engineering problems to solve. But the cost of solving them is falling fast. That’s the pattern we’ve seen with every major technology humanity eventually figured out.

I’m putting SpaceX straight at the top of my watchlist.

But to earn its way into our Disruption Investor portfolio, Starship needs to succeed.

My co-editor Chris Wood and I would also like to see two other conditions met. We talk about those and much more inside the new issue of Disruption Investor, titled The SpaceX IPO Playbook. Members can access it here. 

If you’re not a Disruption Investor member yet, join us here.

Stephen McBride Chief Analyst, RiskHedge