The race to build data centers in orbit just hit a fundamental physics problem. Voyager Technologies CEO Dylan Taylor threw cold water on Elon Musk's ambitious timeline for space-based computing this week, telling CNBC that a two-year deployment would be "aggressive" - and that the industry still hasn't cracked the cooling challenge that could make or break the entire concept. His timing is pointed: Musk just merged SpaceX and xAI in a $1.25 trillion deal partly premised on launching orbital data centers by 2028.
Voyager Technologies just became the voice of caution in the space data center gold rush. CEO Dylan Taylor's blunt assessment - that a two-year timeline for orbital computing facilities would be "aggressive" - lands as a direct counterpoint to the hype surrounding SpaceX and xAI's freshly minted mega-merger.
The problem isn't launch capacity. SpaceX's heavy-lift rockets can haul the hardware to orbit without breaking a sweat. What's missing is far more fundamental: a viable way to keep processors from overheating in the vacuum of space.
"It's counter intuitive, but it's hard to actually cool things in space because there's no medium to transmit hot to cold," Taylor told CNBC's Morgan Brennan in an interview that aired this morning. "So essentially, all heat dissipation has to happen via radiation, which means you need to have a radiator pointing away from the Sun to do that."
That's not a minor engineering wrinkle - it's a physics constraint that could derail the entire orbital data center vision. On Earth, air and liquid cooling systems whisk heat away from processors. In orbit, there's no air, no water, and no convection. Every joule of thermal energy has to radiate into the void, requiring massive radiator arrays that add weight, cost, and complexity.
The timing of Taylor's remarks is hardly coincidental. Just days ago, Tesla CEO Elon Musk announced the combination of SpaceX and xAI in what's being called the largest corporate merger in history at $1.25 trillion. Musk has long championed space-based computing infrastructure as the logical evolution of AI development, arguing that orbital facilities could access limitless solar power and avoid terrestrial real estate constraints.
