SpaceXs, Orbital

SpaceX's Orbital Data Center Bet Meets a Cost Reality Check

Published on 11/02/2026 at 07:02 | Editorial boerse-global.de
Musk expects orbital computing to match terrestrial by late 2028, but Google specialists warn space data centers could cost up to 100 times more.
SpaceX's Orbital Data Center Bet Meets a Cost Reality Check Illustration mit AI erstellt.

Elon Musk spent yesterday defending the efficiency of SpaceX's investments and restating his expectations for data centers in space. For investors, the more pressing question is whether those facilities can pay their way. According to media reports, energy specialists at Google warn that such installations could cost up to 100 times more than terrestrial data centers — a substantial cost caveat hanging over Musk's expansion plans.

The tension cuts to the core of the strategy. SpaceX intends to grow computing capacity not only on the ground but, in future, via satellites as well. Whether that yields an economic edge depends on more than technically available capacity.

Lofty expectations for orbit

Musk expects newly installed computing power in orbit to match its terrestrial equivalent by late 2028 and to surpass it in 2029. That is a forecast, not a milestone already reached. To get there, SpaceX plans up to 1 million Starmind AI satellites, with thousands potentially launching from the end of 2027. The announced scale makes the ambition plain, though on its own it says nothing about achievable returns.

The Google specialists' cost warning therefore challenges a central assumption: a location in space is not automatically more economical. The possible cost gap is likewise an estimate, not a settled calculation for the SpaceX project.

Should investors sell immediately? Or is it worth buying SpaceX?

Analyst Kyle Reidhead puts spending on the targeted capacity build-out at roughly $470 billion over five quarters. That model covers the expansion as a whole and should not be read as a cost estimate for the satellite plans alone. The distinction matters for shareholders. Financing the broader AI build-out and the economics of orbital facilities are connected, but they are not the same question. Rapid business expansion does not answer which slice of the infrastructure will deliver the best long-term returns.

Payback rests on assumptions

Chief financial officer Bret Johnsen says new computing capacity pays for itself in under twelve months at current prices. Musk expects that window to narrow to six months next year. The proviso on current prices is decisive. The statement describes the economics of new capacity under those conditions; it does not prove that future orbital data centers will match those payback periods.

How heavily the long-range math depends on execution is also visible in a model from Sequoia partner Shaun Maguire, which assumes 250 Starship flights in 2028. That assumed launch cadence is a key precondition of his calculation.

Valuing SpaceX therefore cannot rest on the pace of capacity expansion alone. The space strategy only becomes credible once capital needs, technical delivery and financial returns can be brought together. For now, Musk's expectations and the cost warning mark differing assessments of a bill that remains open.

A launch cadence that keeps running

Operationally, the company has not stood still. On Friday, Falcon 9 lifted off from Vandenberg Space Force Base and deployed 27 Starlink satellites into low Earth orbit. Earlier in the week, on Tuesday, SpaceX had launched further satellites from the same site on the Starlink 15-16 mission, with the first stage completing its 26th flight and landing afterwards.

Those are tangible results. Yet the stock market told a different story on Friday: according to media reports, Intel's weaker-than-expected outlook for the fourth quarter weighed on the shares, which fell 4.3% on the day. The report tied the disappointment to doubts about the partnership between Intel and SpaceX. A softer forecast at Intel does not by itself prove the partnership is failing or that the satellite build-out has stumbled — but it can cast doubt on the expectations investors attach to the collaboration. That is where the larger story sits: at SpaceX, spaceflight and the financing of AI infrastructure now share the same valuation sheet.

The $40 billion hardware question

The scale of that second theme emerged on October 26, when Reuters reported an agreement with a subsidiary managed by Apollo Global Management. The entity is to buy AI chips and server hardware and lease them to SpaceX, with total repayment obligations of around $40 billion. The arrangement links access to computing hardware with long-term payment commitments — a more consequential thread behind the nervousness than any single launch. Alongside the question of what SpaceX can technically deliver stands the question of the terms on which the infrastructure it needs becomes available.

SpaceX at a turning point? This analysis reveals what investors need to know now.

SpaceX aimed to close the agreement by year-end, according to Reuters, with leasing arrangements expected to begin in the first half of 2027. Those steps are described as targets and expectations, not as fully executed financing. The gap between agreement and execution is part of the substance here: a successful satellite launch is a completed event, while a sought-after financing close remains a further step. The two cannot be placed on the same footing of certainty.

Progress is not all-clear

What does another successful launch say about the durability of the AI investments? Little, at first. It documents an achievement in spaceflight operations but does not answer the questions on hardware procurement and its financial terms. Equally, a weak Intel outlook does not refute the reported advances in the Starlink build-out.

The sensible yardstick is to examine these developments separately and then bring them together. The recent launches supply operational evidence. The Apollo agreement sketches the planned financial framework for additional AI infrastructure. Intel's outlook shows how sensitive expectations around a partnership can be to news from outside SpaceX. The point of this weekend is not a contest between a good rocket and a weak stock. SpaceX must be measured against several ventures at once — and the next successful launch may confirm progress without settling the open questions on its AI plans.

Ad

SpaceX Stock: New Analysis - 2 November

Fresh SpaceX information released. What's the impact for investors? Our latest independent report examines recent figures and market trends.

Read our updated SpaceX analysis...

Disclaimer...
en | US84615Q1031 | SPACEXS | boerse | 70443393 |