Nvidia is making a staggering $100 billion bet on OpenAI, as the San Francisco-based artificial intelligence lab sets out to build data centers worth “hundreds of billions of dollars” powered by the chipmaker’s AI processors, the companies announced Monday.
At the heart of the partnership is an audacious plan: OpenAI will deploy Nvidia systems requiring 10 gigawatts of power—a scale almost unheard of in the computing world. Nvidia CEO Jensen Huang told CNBC’s Jon Fortt in San Jose that 10 gigawatts translates to 4 to 5 million GPUs, the same number Nvidia expects to ship this year, and “twice as much as last year.”
“This is a giant project,” Huang said in the interview, where he appeared alongside OpenAI CEO Sam Altman and company president Greg Brockman.
The scale of the deal underscores just how tightly intertwined Nvidia and OpenAI have become since ChatGPT ignited the AI frenzy in 2022. Demand for Nvidia’s GPUs surged with the chatbot’s release, and OpenAI remains one of its biggest customers. Now, the chipmaker is going beyond supplier status, with Huang calling the partnership “monumental in size.”
Investments will be staged. A person familiar with the matter said Nvidia’s first $10 billion tranche will be deployed once the first gigawatt of infrastructure is online. The company will continue to invest at then-current valuations as additional gigawatts come online. Nvidia stock jumped almost 4% on the news Monday, instantly adding about $170 billion to its market value, which now hovers near $4.5 trillion.
Industry watchers immediately took note of the circular nature of the deal. “Nvidia invests $100 billion in OpenAI, which then OpenAI turns back and gives it back to Nvidia,” said Bryn Talkington, managing partner at Requisite Capital Management, on CNBC. “I feel like this is going to be very virtuous for Jensen.”
The numbers are dizzying. Huang has previously said that building just one gigawatt of data center capacity costs between $50 billion and $60 billion, with about $35 billion of that going directly to Nvidia chips and systems. The first phase of this new mega-project is expected to go live in late 2026, powered by Nvidia’s next-generation Vera Rubin systems.
For OpenAI, the partnership signals a new chapter. The company now has 700 million active weekly users, and Altman suggested that the infrastructure buildout is key to sustaining momentum. “You should expect a lot from us in the coming months,” he said. “There are three things that OpenAI has to do well: we have to do great AI research, we have to make these products people want to use, and we have to figure out how to do this unprecedented infrastructure challenge.”
Nvidia will remain a “preferred” supplier of chips and networking gear, the companies said, even as rivals like AMD and major cloud providers roll out competing systems. Altman called both Nvidia and Microsoft “passive” investors but also “two of the company’s most critical partners.”
Huang emphasized that the $100 billion commitment is “additive to everything that’s been announced and contracted,” suggesting it won’t alter Nvidia’s existing financial outlook but is instead a bold new layer on top of it.
The deal comes after OpenAI was recently valued at $500 billion in a secondary share sale backed by Microsoft, SoftBank, Thrive Capital, and others. It will also complement infrastructure projects with Microsoft, Oracle, SoftBank, and Stargate.
For Nvidia, the investment extends a recent spree. Last week, it disclosed a $5 billion stake in Intel to collaborate on AI processors, put nearly $700 million into U.K. data center startup Nscale, and reportedly spent over $900 million to hire Enfabrica CEO Rochan Sankar and his team while licensing the startup’s technology.
The sheer scale of Monday’s announcement, however, dwarfs all of that. As Huang framed it, this is nothing less than the construction of the backbone for the next era of artificial intelligence.
Stay ahead in the world of AI, business, and technology by visiting Impact AI News for the latest news and insights that drive global change.
Discover more from Impact AI News
Subscribe to get the latest posts sent to your email.
