Elon Musk has sent shockwaves through the tech industry by announcing a massive leap into energy and compute infrastructure, aiming to deliver between six and ten gigawatts of capacity by 2027. To the casual observer, the numbers seem impossible. Achieving this scale would require capital expenditures potentially reaching five hundred billion dollars, a figure usually reserved for the world’s largest cloud giants rather than a rocket company. However, analysts suggest this ambition is grounded in a stark reality: high scale combined with immediate availability is currently one of the rarest and most valuable commodities in the artificial intelligence race.
The economic engine driving this expansion is the staggering profitability of AI inference. While building these clusters is expensive, the revenue generated by serving tokens via frontier models like those from OpenAI and Anthropic can exceed one hundred billion dollars per gigawatt annually. Because SpaceX intends to bring these facilities online far faster than traditional providers—offering lead times of just three to five months—they can command premium pricing that allows them to recoup their investment in under a year. This aggressive strategy could realistically propel SpaceX toward an annual recurring revenue of three hundred billion dollars by the end of 2027.
Microsoft is poised to be the primary beneficiary and partner in this land grab for power. By leveraging its relationship with OpenAI, Microsoft can capture immense margins without bearing the initial brunt of training costs. Reports indicate that Microsoft has already signed ten gigawatts of contracts this year to bridge a critical capacity gap heading into 2028. Given that these agreements often include flexible cancellation policies, there is little balance sheet risk for Microsoft, but an enormous upside that could potentially double Azure’s revenue growth rate almost overnight.
To fund such an astronomical rollout without the existing treasury of a legacy hyperscaler, SpaceX is expected to lean on strategic partnerships and unconventional financing. This includes heavy support from Nvidia through vendor financing, explaining Musk’s recent pivot toward exclusivity with their chips over competitors like AMD or Google’s TPUs. By combining Nvidia’s hardware backing with record-breaking rental rates for rapid-deployment compute, SpaceX isn’t just launching satellites anymore; it is positioning itself as the indispensable backbone of the global AI economy.