/

Real-World Experience

Real-World Experience

AI Training in Orbit: Why the Future of Data Centres Lies in Space

AI Training in Orbit: Why the Future of Data Centres Lies in Space

AI Training in Orbit: Why the Future of Data Centres Lies in Space

SpaceX and Starcloud are powering the AI future from orbit: the first Large Language Model is being trained directly in space. This article analyses the economic drivers behind orbital supercomputers – from unlimited solar energy to the elimination of electricity and cooling costs – and shows why the future always sounds like science fiction to decision-makers at first.

SpaceX and Starcloud are powering the AI future from orbit: the first Large Language Model is being trained directly in space. This article analyses the economic drivers behind orbital supercomputers – from unlimited solar energy to the elimination of electricity and cooling costs – and shows why the future always sounds like science fiction to decision-makers at first.

Marcel Pesch

•

The first Large Language Model is currently being trained directly in space.

When you read this sentence, at first it sounds like the script of a Hollywood blockbuster. But we are not talking about science fiction here, but rather about the real infrastructure strategy of the leading space and tech pioneers of our time.

Starcloud is training the first global AI model on satellites in orbit. At the same time, SpaceX's plans for a gigantic mega-project in space are solidifying:

  • 5 gigawatts of cumulative computing power in Earth orbit.

  • 16 kilometres of extended solar panels for clean energy generation.

  • A planned $1.5 trillion IPO to finance precisely this orbital supercomputer architecture.

Anyone who thinks this is an absurd idea is ignoring the hard physical and economic limits that terrestrial data centres on Earth are currently hitting.

The 3 Levers of Orbital AI Infrastructure

Why are tech giants moving AI training into space? The reason lies in the massive resource bottlenecks on the ground:

  1. No electricity costs: On Earth, modern AI infrastructures consume the equivalent of major cities. In orbit, continuous solar radiation outside the atmosphere provides unlimited, completely free solar energy.

  1. Free cooling: The cooling requirements of terrestrial data centres consume billions of litres of water and massive amounts of electricity. In space, the freezing cold environment of the vacuum serves as a natural, inexhaustible heat sink.

  1. No land-use conflicts: No lengthy approval processes for building plots, no resident protests, and no overloading of local power grids.

What sounds crazy at first glance is, upon closer inspection, the only logical scaling response to the exponential hunger for AI computing power.

The "Crazy" Paradox: How the Present Emerges

In business, we tend to dismiss groundbreaking developments as unrealistic at first. However, a look at recent technological history reveals a clear pattern:

  • 10 years ago, the idea of carrying a smartphone in your pocket that could solve logical tasks more complex than a human could sounded like pure fantasy.

  • 5 years ago, a generative chatbot that writes specialist books and programming code at the push of a button sounded completely abstract.

  • 2 years ago, an autonomous AI agent that independently manages business processes and makes decisions sounded like a fictional experiment.

The future always sounds crazy until it gradually becomes our everyday reality.

Leaders and decision-makers who want to steer successfully in the age of AI must no longer evaluate developments based on today's status quo. Anyone who only believes what they already know to be possible is always two steps behind technological reality.

Conclusion: Opening the Horizon for the Impossible

The shift of AI data centres to orbit by SpaceX and Starcloud is the ultimate proof of how rapidly benchmarks are shifting. Today's limits – whether computing power, energy consumption, or scalability – are tomorrow's solved problems.

At academy4.ai, we prepare decision-makers to think of technology in exponential terms. Not to chase every hype, but to recognise future market opportunities early and exploit them strategically.

What sounds completely crazy to you today – but will be part of our absolute everyday life in 5 years?

MMAI® EXPEDITION

Master AI.

Remain Human.

Become a certified AI Leader in 3 months. Hands-on, with mountain guide support.

Start expedition →

Your summit is waiting.

Master AI. Remain Human.

Ready for your ascent?

The next expedition starts soon. Spaces are limited (max. 6–7 per rope team). Become a summit conqueror.

Next rope team starts in
00
Days
00
hrs.
00
Min.
Free route advice