Why SpaceX is moving into gas turbine manufacturing
Subscribe to our free newsletter today to keep up to date with the latest manufacturing news.
SpaceX is preparing to manufacture gas turbine blades and vanes in Texas as Elon Musk looks for ways to secure more power for AI infrastructure.
The company is hiring workers for a blades and vanes foundry in Bastrop, Texas. The facility would produce parts used in industrial gas turbines, which can generate electricity for data centers and other large sites.
Musk has said producing the components in-house could cut as much as 18 months from the time needed to bring new gas turbines online.
That claim has not been independently proven. But the move points to a wider problem facing the AI industry. Companies can buy more chips and build more servers, but those systems still need large amounts of electricity.
As data center construction grows, access to power equipment is becoming a major constraint.
AI growth is putting pressure on electricity supply
Data centers already account for a growing share of electricity demand in the US.
The International Energy Agency estimates that US data centers consumed about 180 terawatt-hours of electricity in 2024. It expects their use to increase by roughly another 240 terawatt-hours by 2030.
The agency also expects data centers to account for about half of the increase in US electricity demand between 2025 and 2030.
Globally, data center electricity use is projected to rise from about 485 terawatt-hours in 2025 to around 950 terawatt-hours in 2030. Much of that growth will come from AI.
This is creating a new set of industrial bottlenecks. Semiconductor supply remains important, but companies also need transformers, substations, transmission links, backup systems and generation equipment.
Gas turbines are one option because they can provide large amounts of steady power. They can also be installed near major data center sites, reducing dependence on new transmission infrastructure.
However, turbine supply has become tight.
Large industrial gas turbines require complex components that are difficult to produce. Turbine blades must operate under extreme heat and mechanical stress. Their manufacture depends on specialist casting methods, advanced alloys, coatings and strict quality controls.
Those requirements limit the number of suppliers capable of producing them at scale.
For AI companies trying to add computing capacity quickly, long manufacturing lead times can delay an entire project.
SpaceX is applying its manufacturing model to energy
SpaceX has built much of its business around vertical integration.
Rather than relying entirely on outside suppliers, the company manufactures many components itself. This has allowed SpaceX to make design changes quickly and reduce its reliance on traditional aerospace supply chains.
The proposed turbine foundry suggests that SpaceX is applying the same approach to power infrastructure.
Its hiring materials describe power generation as a possible limit on global AI adoption. The company is seeking workers with experience in investment casting and turbine component production.
That does not mean SpaceX is about to become a major gas turbine supplier.
Building the foundry is only one part of the process. Turbine blades require extensive engineering, testing and certification. Establishing consistent production at high volumes can take years.
Analysts at Jefferies have also suggested that commercial SpaceX turbine products could still be several years away.
There is also an important difference between manufacturing turbine components and building a complete turbine system. SpaceX may be able to shorten part of the supply chain without replacing established turbine makers.
Still, the strategy matters because it shows how far AI infrastructure companies may be willing to go to secure electricity.
Technology companies have already signed long-term deals for nuclear, renewable and natural gas power. Some data center developers are also building generation close to their facilities.
SpaceX is extending that approach by moving into the manufacturing process behind the equipment used to produce power.
Power access could shape the next phase of AI competition
The AI industry has spent much of the past several years focused on computing hardware.
That focus is widening as companies face limits elsewhere in the infrastructure chain.
They need land, power connections, transformers, cooling systems and generation capacity before new processors can begin operating.
For manufacturers, this could create demand from companies that previously had little involvement in the energy equipment market.
It could also change the relationship between data center operators and suppliers.
If turbine lead times remain long, larger technology companies may seek joint ventures, long-term capacity agreements or direct investment in manufacturing. Some may choose to produce selected components themselves.
That approach will not work for every company. Heavy industrial manufacturing requires skills, capital and production systems that differ greatly from software development.
SpaceX is an unusual case because it already operates large manufacturing facilities and has experience bringing complex production processes in-house.
Even so, its turbine project illustrates a wider shift in the AI market.
The race to build more computing capacity increasingly depends on access to physical infrastructure. Chips remain central, but they cannot operate without reliable power.
That places greater value on the companies that manufacture equipment for electricity generation.
For turbine makers, utilities and industrial suppliers, AI demand could become an important source of investment. For data center operators, energy planning may need to start much earlier in project development.
SpaceX’s move into turbine components shows why. The next limit on AI growth may not be inside the server rack. It may be on the factory floor.
Source