The Sequence Opinion #888: Everything You Need to Know About the AI in Space Race
Use cases, key platforms, architectures, challenges and more.
TL;DR
- Space is becoming a new and competitive frontier for AI.
- The current AI frontier is defined by energy scarcity (grid capacity, cooling, land, permits).
- Low Earth orbit offers effectively unmetered energy and no jurisdictional limits, making it attractive for AI development.
- Space is no longer a hypothetical frontier but contested economic territory with tangible investments and operations.
- Trillion-dollar companies, hyperscalers, chipmakers, nation-states, and startups are actively involved.
- The first large language model trained in space was nanoGPT, trained on Shakespeare aboard a satellite in December 2025.
- The core thesis is that compute has become an energy problem, and space is a potential energy solution.