Nvidia, a maker of chips for artificial intelligence, is investing $100 billion in ChatGPT parent OpenAI to fund a massive expansion of data centers using their combined technology.
The deal aims to expand the systems and real estate needed to power a not-so-far-away future in which people and economies rely far more on AI than they do now, the companies said in statements.
Through the deal, Nvidia and OpenAI are expected to build new data centers using millions of Nvidia’s chips, involving 10 gigawatts of power. To give some perspective, 1 gigawatt can power about 750,000 homes at the same time.
With the U.S. data center market currently estimated at 40 gigawatts, this single deal could boost data center inventory by 25%, according to Howard Huang, a market intelligence analyst at Avison Young’s San Francisco office who tracks the sector.
Nvidia and OpenAI are already involved in some of the largest data center developments taking hold across the country. Stargate — a program backed by OpenAI, Oracle and other AI industry leaders — aims to spend up to $500 billion on data centers and other infrastructure in the U.S. that will be powered by Nvidia’s chips.
The scope of this latest deal between Nvidia and OpenAI shows that “the AI data center race is still in its early stages, with tremendous momentum,” Avison Young’s Huang said.
Nvidia CEO Jensen Huang, who expects global AI infrastructure spending on real estate and other support systems to hit $4 trillion by the end of the decade, called the latest partnership “the next leap forward to power the next era of intelligence.”
This artificial intelligence potential is threatened by headwinds to data center development, including power and water constraints and high construction costs. Nvidia is facing its own challenges tied to those headwinds. The second quarter marked Nvidia’s slowest growth since 2023, when the generative AI boom began driving the company’s results.
Data center development boom
OpenAI, Amazon, Google, Meta and Microsoft together have announced plans to spend more than $300 billion combined on these specialized facilities that house powerful computer servers, storage systems and networking equipment.
“Rivals like Meta, X.AI, Google and Microsoft are all vying to outspend one another to gain an edge in AI,” Avison Young’s Huang told CoStar News. “I expect to see a wave of similar partnerships and large-scale investments announced in the near future.”
Microsoft projected that it would spend a record $30 billion in the current quarter alone as it races to catch up with a backlog of demand for cloud and other offerings such as Microsoft 365 Copilot, an AI assistant. Seattle-based Amazon nearly doubled its capital spending to over $32 billion in the second quarter, much of it on data center developments, and expects to spend up to $100 billion this year on investments related to artificial intelligence for its Amazon Web Services cloud computing service.
“I think it’s pretty safe to say that we’re all going to be consuming more data next year than we did the year before,” Andrew Batson, head of data center research for Chicago-based commercial real estate services firm JLL, told CoStar News in a recent interview.
The technology industry could invest up to $1 trillion in the next five years to build facilities across North America, according to a recent report by JLL.
Real estate push
Nvidia has expanded from a little-known Silicon Valley semiconductor outfit to the world’s most valuable publicly traded company at the forefront of the AI boom in part by investing in companies that depend on its products.
Last week, the firm said it was investing $5 billion in struggling chipmaking rival Intel, as the two Silicon Valley firms collaborate on technologies geared to high-demand data centers. The company also recently announced more than $2.5 billion in investments in AI infrastructure projects in the United Kingdom.
Nvidia has spent close to $1 billion since last year on acquiring properties near its Santa Clara, California, headquarters.
San Francisco-based OpenAI, meanwhile, has been having an aggressive growth spurt of its own in recent years. The firm has finalized two major lease deals in less than two years, stretching its regional footprint by nearly 800,000 square feet. Executives have said the firm ultimately plans to double its current 2,000-person workforce.
The firm said the deal “complements the deep work” OpenAI and Nvidia are already doing with other companies such as Microsoft, Oracle, SoftBank and Stargate to “advance its mission to build artificial general intelligence that benefits all of humanity.”