OpenAI, on September 23, announced five new AI data centre sites, along with Oracle and SoftBank, in the United States.
The announcement is part of the Stargate Project, the $500 billion initiative to build AI infrastructure across the US, led by OpenAI and SoftBank in partnership with Oracle, MGX, Arm, Microsoft and NVIDIA.
OpenAI states that the combined capacity from these five new sites brings Stargate to nearly 7 gigawatts (GW) of planned capacity, in addition to ongoing projects with CoreWeave.
“This puts us on a clear path to securing the full $500 billion, 10-gigawatt commitment we announced in January by the end of 2025, ahead of schedule,” the company said.
In July, OpenAI and Oracle signed a deal valued at over $300 billion to develop up to 4.5 GW of Stargate capacity within the next five years. This partnership powers three new sites located across the US, which are expected to deliver 5.5 GW of capacity, including a potential expansion of 600 megawatts (MW) near the ‘flagship’ Stargate site in Abilene, Texas.
The three sites are located in Shackelford County, Texas; Doña Ana County, New Mexico; and a site in the Midwest, which is expected to be announced soon.
On the other hand, the other two Stargate sites can scale to 1.5 GW of power over the next 18 months. These sites will be developed through a partnership with SoftBank, which is another major partner in the Stargate Project.
These two sites are located in Lordstown, Ohio and Milam County, Texas.
Recently, OpenAI’s CEO, Sam Altman, penned a blog on his personal website, outlining the trajectory of AI and computing power. “Our vision is simple: we want to create a factory that can produce a gigawatt of new AI infrastructure every week,” he said, adding that the execution of this will be “extremely difficult” and could take years to get to the milestone.
“Maybe with 10 gigawatts of compute, AI can figure out how to cure cancer. Or with 10 gigawatts of compute, AI can figure out how to provide customised tutoring to every student on earth,” Altman added.
A few days ago, OpenAI and NVIDIA signed a letter of intent for a partnership to deploy at least 10 gigawatts of NVIDIA systems for OpenAI’s next-generation AI infrastructure.
The agreement includes NVIDIA investing up to $100 billion in OpenAI, contingent upon the deployment of each gigawatt of systems.
The first gigawatt of capacity is scheduled for deployment in the second half of 2026 on NVIDIA’s Vera Rubin platform.
