Starcloud wants to put Bitcoin mining hardware in space, but the more important story is not the coin. It is how far companies are willing to go to find cheaper power for compute.
Bitcoin mining has always followed the cheapest available energy. It moved from laptops to warehouses, from hydro-rich regions to stranded gas fields, from quiet industrial towns to sites built around power contracts. Starcloud is now pushing that logic into orbit, with a plan to fly Bitcoin mining ASICs on its second spacecraft and test whether proof-of-work can run above the grid entirely.
The company behind the plan is Starcloud, the Redmond, Washington startup formerly known as Lumen Orbit. It is building orbital data centers for AI and other compute-heavy workloads, and it already has one meaningful proof point. Starcloud-1 launched in November 2025 with an Nvidia H100 GPU on board, making it one of the first serious demonstrations of data-center-grade compute in low Earth orbit.
Now the pitch has widened. According to Cointelegraph, CEO Philip Johnston said Starcloud plans to begin mining Bitcoin from space when its second spacecraft launches, using application-specific integrated circuit miners rather than GPUs. His argument is straightforward: ASICs are much cheaper per watt than the AI chips that dominate cloud computing, with Johnston comparing a roughly $1,000 1-kilowatt ASIC to a roughly $30,000 1-kilowatt AI GPU.
The claimed energy source is orbital solar power. In Starcloud's broader model, satellites use sunlight without the interruptions and land constraints faced by terrestrial solar projects, while the cold environment of space helps with heat rejection through radiative cooling. That matters because mining margins are brutally sensitive to electricity costs. A miner with low-cost power can survive a Bitcoin drawdown or a difficulty spike far longer than one paying normal industrial rates.
There is an obvious reason this story caught attention. Bitcoin mining has spent years defending its power use against regulators, utilities and local communities. In Texas, miners have been praised for flexible demand and criticized for grid strain. In New York, policy fights have centered on fossil-fuel plants repurposed for mining. In Kazakhstan, a mining boom ran into power shortages and a government crackdown. Space-based mining sounds like an escape hatch from all of that, because it shifts the load away from the terrestrial grid.
But it is still a planned experiment, not an operating mining business. Starcloud has not shown that any Bitcoin has been mined in orbit. It has not disclosed the expected hash capacity of the payload, the number or model of ASICs, or the economics after launch cost, satellite lifetime, communications, radiation risk and maintenance constraints are included. For Bitcoin itself, the initial contribution would almost certainly be negligible compared with a global network measured in hundreds of exahashes per second.
That does not make the test meaningless. It makes it an infrastructure signal. If a smallsat can host ASICs, keep them powered, manage heat, maintain network connectivity and submit valid work to a mining pool, Starcloud will have shown that orbital compute can handle a simple but unforgiving workload. Bitcoin mining is not elegant computing, but it is continuous, measurable and financially transparent. That makes it a useful stress test.
AI may be the bigger prize
The more valuable question is whether orbital data centers can serve AI demand, not whether one satellite can win a Bitcoin block. Starcloud-2 is described by the company as a commercial mission with a GPU cluster, persistent storage, around-the-clock access and proprietary thermal and power systems in a smallsat form factor. The company says it is designed for sun-synchronous orbit and full operation by 2027, which suggests the Bitcoin plan sits inside a longer timeline for orbital cloud infrastructure.
That matters because the AI data center market is running into the same constraints that shaped Bitcoin mining, only at a much larger institutional scale. Power availability, permitting, cooling, land, water use and grid interconnection delays are now board-level issues for hyperscalers and AI labs. If launch costs keep falling, space becomes less of a stunt and more of a strange but logical extension of the energy arbitrage trade.
On Earth, Bitcoin miners are already trying to move in that direction. Companies with large power positions, grid relationships and data center sites are increasingly pitching themselves as high-performance computing providers for AI customers. The reason is simple. AI workloads can produce steadier revenue than mining, while using some of the same power infrastructure. TeraWulf, Core Scientific, Cipher Mining and others have all been watched through that lens by investors looking for miners that can become compute landlords.
Starcloud is approaching the same convergence from the opposite direction. Instead of taking a mining site and asking whether it can host AI servers, it is taking an orbital AI data center thesis and asking whether Bitcoin ASICs are the easiest workload to prove first. That is a clever framing, but it also raises the bar. A promotional clip can make space mining sound inevitable. A working satellite has to survive radiation, thermal cycles, downlink limits and a hardware market where ASICs age quickly.
The practical takeaway is that Bitcoin mining in space is not about Bitcoin becoming interplanetary money tomorrow. It is about compute chasing energy to increasingly unusual places. If Starcloud mines even a small amount of Bitcoin from orbit, the market will treat it as a milestone. If it proves that orbital power and cooling can support commercial compute at scale, the much larger battle will be over AI infrastructure, not block rewards.
Also read: SafePal faces a trust test after a reported customer data leak • Coinbase outage puts crypto infrastructure back under pressure • Japan's bond market tokenization plan could turn blockchain into financial plumbing
