amynicole – OpenAI has announced a major partnership with Oracle to build 4.5 gigawatts of new data center capacity across the United States. The infrastructure will support the ever-growing compute demands of OpenAI’s large language models. Although the companies have not yet confirmed specific locations, Bloomberg reports that Texas, Michigan, Wisconsin, and Wyoming are among the states under consideration.
Read More : Nokia Turns to Reddit to Find Smartphone Manufacturing Partner
This massive buildout is part of OpenAI’s Stargate Project, which aims to significantly expand U.S. AI capacity. When combined with its existing 1-gigawatt facility in Abilene, Texas, OpenAI expects its total power capacity to exceed five gigawatts. That would support more than two million AI chips once construction is complete.
Oracle recently began delivering GB200 server racks to the Abilene site, which has already started operations. These hardware installations will play a central role in powering OpenAI’s next-generation AI systems. The company estimates the initiative will create more than 100,000 jobs. Though most of these are expected to be temporary construction roles rather than long-term positions.
The announcement marks a significant step toward OpenAI’s larger goal of investing $500 billion in U.S.-based AI infrastructure. That broader vision includes building 10 gigawatts of compute power over the next four years. OpenAI is partnering with Oracle and SoftBank on this expansion. Though SoftBank is not involved in financing this specific phase of the project.
AI Industry Races to Secure Energy and Compute at Scale
OpenAI’s investment comes amid a broader industry push by major tech companies to secure energy and data center resources for AI development. Google recently committed £3 billion to secure 3,000 megawatts of hydroelectric power. Microsoft is planning to revive the long-idled nuclear facilities at Three Mile Island to power its AI ambitions. These moves highlight how energy and compute resources have become central to competition in AI.
OpenAI’s Stargate Project is part of this trend and demonstrates the scale required to train and operate large language models. In its press release, OpenAI described Stargate as an “ambitious undertaking” backed by strong public and private support, including encouragement from the White House. The administration recognizes AI infrastructure as essential for innovation, economic growth, and national competitiveness.
Read More : Microsoft Says Chinese Hackers Breached Its Servers
Several U.S. states are now offering aggressive tax incentives to attract data center projects. This competition is helping OpenAI and its partners identify favorable locations for rapid construction and deployment. As demand for generative AI continues to grow, OpenAI’s collaboration with Oracle is positioned to play a leading role in shaping the future of compute infrastructure. The success of these data centers will be critical not only to OpenAI’s future models but also to maintaining U.S. leadership in AI development.
