THE ORBITAL AI FACTORY ROADMAP
The 3-Phase Scaling Path
An Orbital AI Factory scales in three clear steps. We start with manufacturable 100 kW-class nodes to validate economics and operations in LEO, then scale to megawatt-class factory nodes, and finally unlock the next cost curve with off-Earth manufacturing.
Phase 1 — 100 kW Edge Nodes
Localized AI near sensors
The first truly scalable class of orbital AI infrastructure.
Optimized for low mass, high packing efficiency, and repeatable manufacturing. Uses liquid-cooled AI racks with heat rejected through low-mass deployable radiators. Primary model: process in orbit, beam back results.
Why it matters: This is the fastest path to real, commercial-scale orbital AI.
Phase 2 — 1 MW Factory Nodes
Our Reference Design
Serviceable, upgradeable orbital data center nodes
The mature "regional backbone" of an Orbital AI Factory. ROSA-class mega-wings, large deployable radiators, modular compute pods. High-bandwidth laser networking enables synchronized workloads across multiple nodes. Built for robotic servicing and multi-generation upgrades.
Why it matters: This is where orbital AI becomes a true data center architecture.
Phase 3 — Off-Earth Manufacturing
The multi-TW scaling horizon
Lunar/cislunar manufacturing of structures, arrays, and radiators.
Potential mass-driver logistics dramatically reduce Earth launch dependence. Enables orders-of-magnitude expansion of orbital compute capacity.
Why it matters: This is the long-term path to civilization-scale infrastructure.