SpaceX is betting on manufacturing one of the most difficult gas turbine components at its own facilities, in an effort to reduce the time needed to bring online the energy sources required by AI data centers. Elon Musk confirmed on August 30, 2026, that a secret foundry the company is building in Bastrop, Texas, will cast turbine blades and vanes, saying this could accelerate the commissioning of gas turbines by up to 18 months.
Musk’s confirmation came after a report by The Information, based on job postings that explicitly referred to a facility for casting “blades and vanes,” as well as information that SpaceX purchased approximately 830 acres near its existing Starlink plant in Bastrop between March and June. Musk said on X that SpaceX and Tesla are each building 100 gigawatts of annual solar power production capacity as quickly as possible, but added that natural gas would remain necessary for years to support solar power and bring it online.
A New Bottleneck Behind AI Expansion
The development is not linked solely to a shortage of AI chips. Data centers require enormous amounts of electricity, while power grids cannot always provide it as quickly as new computing projects demand. The International Energy Agency estimates that global data center electricity consumption could nearly double by 2030. GE Vernova, one of the gas turbine manufacturers, also says its production capacity is almost sold out through 2030, largely because of demand generated by AI infrastructure.
For this reason, building gas-fired plants near data centers has become an increasingly used alternative to waiting for grid expansion. Companies including Amazon, Google, Meta, OpenAI, and Microsoft are adopting this approach after focusing for years on wind and solar power. In this context, manufacturing turbine components in-house could give SpaceXAI a way to bypass part of the supply chain on which other data center developers depend.
Why Are Turbine Blades Difficult to Manufacture?
The blades located in the hottest section of gas turbines operate at temperatures ranging from 3,000 to 3,600 degrees Fahrenheit, approximately 800 degrees higher than the melting point of the metal alloy from which they are made. This is possible because of internal cooling channels, thermal-barrier coatings, and the precise casting method used for each blade.
Manufacturing also requires producing each blade as a single, continuous crystal that grows slowly inside a vacuum furnace, to avoid microscopic joints that could allow ordinary metal to crack under stress. The challenge increases when the blades are intended for power-plant turbines because they are larger than aircraft-engine blades. According to the article, only four companies worldwide have mastered this process at industrial scale, and all of them are currently operating at the limits of their production capacity.
If SpaceX succeeds in building this capability, an entity associated with Musk would possess a manufacturing skill on which other AI infrastructure companies depend through a very limited number of suppliers. That could create an advantage difficult for a competitor with funding alone, but without comparable manufacturing expertise, to replicate quickly. However, this possibility depends on engineering and manufacturing success that has not yet been achieved; casting large, defect-free blades is not an automatic result of establishing a new foundry.
What Changes in Practice?
The potential change is a reduction in the time required to deliver power to data centers, not a complete solution to the energy problem. The foundry could ease a specific bottleneck in turbine manufacturing, but it would not eliminate the need for fuel, gas infrastructure, permits, or emissions controls. Accelerating turbine deployment could also increase environmental pressure in areas hosting data centers.
In Memphis, where SpaceXAI has operated gas turbines to power Colossus data centers since 2024, the NAACP has repeatedly accused the company of operating the turbines without the permits or pollution controls required by federal law. The organization says these turbines release compounds that help form smog and hazardous substances such as formaldehyde, pollutants associated with asthma, respiratory diseases, and some types of cancer. Researchers at the University of Memphis also said, in an analysis they described as limited, that air pollution had increased “slightly” because of the data center.
The issue is not limited to Memphis. In Virginia’s “Data Center Alley,” a study commissioned by the Piedmont Environmental Council and using the U.S. Environmental Protection Agency’s COBRA model concluded that emissions from a single facility containing eight full-time gas turbines could affect more than 2.5 million people across several counties. The study estimated between 3.4 and 6.5 additional premature deaths annually, along with annual health damages ranging from $53 million to $99 million, with the greatest impact falling on marginalized communities.
certi.news’s Analysis
The move reveals a practical trade-off in AI expansion: manufacturing integration can accelerate the provision of computing capacity, but it may shift the cost of the bottleneck from the turbine supply chain to communities surrounding power plants. The available facts establish the existence of a foundry project and statements about acceleration of up to 18 months, but they do not yet establish that SpaceX has succeeded in producing blades at industrial scale or that this acceleration will actually occur.
The open question is whether faster energy deployment will be accompanied by adequate permits and emissions controls, and who will bear its health effects in nearby areas. The project therefore should not be evaluated only as a solution to data centers’ electricity shortage; its technical success could make the debate over pollution and regulation more urgent, not less.