Elon Musk says SpaceX is “doing in-house casting” for blades and vanes, which can accelerate natural gas turbines “coming online by up to 18 months”
Elon Musk intends to bypass the power supply chain for AI data centers in a way others assumed was impossible, by making highly complex components himself.
Context & Ripple Effects
The move follows signs that SpaceX was developing a Bastrop foundry for large-turbine blades, while xAI had already disclosed plans to replace 69 Colossus-area turbines with a 1.2 GW natural-gas plant by July 2027. It turns a suspected manufacturing effort into a stated attempt to remove a specific equipment constraint from AI-data-center power buildouts.
Earlier turbine purchases for xAI's data centers showed how heavily its expansion depended on available generation equipment. In-house casting brings a critical, difficult-to-source component under SpaceX's control rather than leaving deployment timing entirely to outside suppliers.
First-order effects
- SpaceX can supply blades and vanes for its own turbine program, with Musk claiming the change can pull natural-gas generation online by as much as 18 months.
- xAI's planned Colossus power transition gains a manufacturing path less exposed to the component lead times that accompany its shift from mobile turbines to a 1.2 GW plant.
Second-order effects
- SpaceX's power buildout becomes more vertically integrated: its foundry capability can determine part of xAI's generation schedule alongside turbine procurement.
- Other AI-data-center developers remain dependent on external turbine-component supply, making SpaceX's control of casting a potential execution advantage rather than merely a purchasing advantage.
Third-order effects
- If SpaceX can repeat the claimed schedule improvement, AI infrastructure builders may treat specialized power-equipment manufacturing as a strategic part of the stack, not a supplier relationship to outsource.
- The pattern would extend the race to deploy AI capacity from securing GPUs and sites into controlling the industrial inputs that govern when generation can be commissioned.
The trend: AI infrastructure is driving vertically integrated approaches to power supply as developers seek to bypass equipment bottlenecks that delay data-center capacity.