Nvidia Commits $100B to Power OpenAI’s Next Wave of AI Growth

Nvidia has struck a landmark deal with OpenAI, pledging up to $100 billion to build massive “AI factories” — next-generation data centers designed to train and run advanced artificial intelligence models like ChatGPT.

The investment marks the largest private commitment yet from a tech company, and reflects CEO Jensen Huang’s determination to keep Nvidia at the center of the AI boom. The chipmaker, valued at over $4 trillion, already dominates the market for AI processors and infrastructure.

While analysts acknowledge Nvidia has the cash flow to fund the project, questions remain: how quickly can these facilities be built, and where will the necessary energy come from? The International Energy Agency estimates that 10 gigawatts of AI data centers — the scale Nvidia and OpenAI are targeting — would consume as much electricity annually as 10 million U.S. homes.

The Deal’s Structure

• Some worry about the circular nature of the deal, since OpenAI will largely be leasing Nvidia’s chips through the investment itself. Still, Wall Street sees the project potentially driving hundreds of billions in future revenue for Nvidia, reinforcing its dominance at a time when rivals like Google, Meta, Amazon, and Microsoft are developing their own custom AI processors.

• The partnership underscores OpenAI’s urgent need for computing power. With over 700 million weekly users on ChatGPT, CEO Sam Altman has argued that AI innovation now depends more on infrastructure than on new ideas. By tying OpenAI even more closely to Nvidia’s hardware and software ecosystem, Huang is effectively betting that control over compute capacity will shape the future of AI.

A Long-Running Partnership

• The move continues a relationship that began in 2016, when Nvidia delivered what Huang called the world’s first AI supercomputer to a then-little-known OpenAI. Now, nearly a decade later, their partnership has scaled into one of the largest technology investments in history.