- DOE selects 16 federal sites for AI data centre development.
- Initiative aims to co-locate data centres with new energy infrastructure.
What happened: DOE announces potential sites for AI data centres
The U.S. Department of Energy (DOE) has identified 16 federal sites suitable for developing artificial intelligence (AI) data centres. This initiative seeks to position the United States as a leader in AI technology and reduce energy costs by co-locating data centres with new energy infrastructure.
The selected sites include national laboratories and facilities such as the Idaho National Laboratory, Paducah Gaseous Diffusion Plant, and Los Alamos National Laboratory. These locations offer existing energy infrastructure and the potential to expedite permitting for new energy generation, including nuclear reactors.
The DOE has issued a Request for Information (RFI) to gather input from data centre developers, energy developers, and the public. The goal is to develop partnerships that will facilitate the construction of AI infrastructure at these sites, with operations targeted to begin by the end of 2027.
Also read: Data centres turn to PPAs for cost stability
Also read: The promising future of underwater data centres: Why aren’t we diving In?
Why it’s important
The DOE’s identification of 16 federal sites for AI data centre development is a strategic move to enhance the United States’ position in the global AI landscape. By leveraging existing energy infrastructure at these sites, the initiative aims to streamline the development process and reduce associated costs.
Co-locating data centres with energy generation facilities, including potential nuclear reactors, could provide a reliable power supply to support the substantial energy demands of AI operations. This approach aligns with efforts to promote clean energy use and manage the environmental impact of expanding data centre infrastructure.
The DOE’s RFI seeks to foster public-private partnerships, encouraging collaboration between government agencies and industry stakeholders. This collaborative approach aims to accelerate the deployment of AI data centres, with the goal of commencing operations by late 2027.
