The OpenAI and Microsoft partnership is growing, and executives from both companies are reportedly developing plans to build a U.S. data center that could cost as much as $100 billion.
As reported by “Information”, citing anonymous sources the data center would house a supercomputer composed of millions of AI chipsand is referred to as the “Stargate”. People he talked to OpenAI CEO Sam Altman and who have reviewed Microsoft’s cost estimates, told The Information that the project could reach $100 billion and that Microsoft would likely finance it. According to The Information, the proposed project costs about 100 times more than some of the largest data centers today, and Microsoft executives want to launch it as early as 2028.
Stargate will be the largest in a series of five installations that Microsoft and OpenAI want to roll out over the next six years, while OpenAI’s next AI update is reported to be available early next year. Meanwhile, Microsoft is developing a supercomputer for Phase 4, which is expected to launch in 2026, and the two companies are in Phase 3 of the plan.
“Microsoft has demonstrated its ability to build a pioneering artificial intelligence infrastructure for training and deploying the world’s leading artificial intelligence models,” a Microsoft spokesman said in a statement. “We are always planning for the next generation of infrastructure innovations necessary to continue to push the boundaries of what AI can do.”
OpenAI did not immediately respond to a request for comment.
Microsoft has made a ‘multi-year, multi-billion dollar investment’ at OpenAI, the creator of ChatGPT, in January last year. This started a fierce and brisk AI race. But the partnership between the two companies fell through control by the EU AND UK Regulators.
And Tesla CEO Elon Musk, co-founder and former director of OpenAI, filed a lawsuit in connection with the company.
