Nvidia is in talks to invest up to $30 billion in OpenAI as part of a massive funding round that could put the AI company at roughly $730 billion pre-money, CNBC reported, citing people familiar with the discussions. The terms aren’t final, and the details could still change.
Importantly, the potential $30 billion commitment would be separate from the headline-grabbing infrastructure framework OpenAI and Nvidia announced in September 2025—an agreement that outlined a multi-year plan to bring new supercomputing capacity online with progressive investment tied to deployment.
Recommended AI dev read: “Hands-On Machine Learning” (Aurélien Géron)
Check price on AmazonThe Financial Times first reported the possible Nvidia investment, with CNBC later confirming discussions. Nvidia declined to comment.
Separate from the $100B infrastructure framework
The new talks come as the status of OpenAI and Nvidia’s previously announced “up to $100 billion” infrastructure framework has been under scrutiny. That September 2025 announcement described a structure where Nvidia would invest progressively as new GPU supercomputing capacity—measured in gigawatts—was deployed.
But in January, The Wall Street Journal reported the deal was “on ice,” fueling questions about whether the relationship had cooled or whether the original framework was simply evolving behind the scenes.
Executives try to tamp down the noise
OpenAI CEO Sam Altman has publicly downplayed the drama, posting on X that OpenAI “loves working with Nvidia” and saying he didn’t understand where the “insanity” was coming from.
Nvidia CEO Jensen Huang also addressed the chatter in a CNBC interview earlier this month, saying there was “no question” Nvidia would participate in OpenAI’s next funding round—comments that now look like a preview of the scale OpenAI may be aiming for.
A mega-round that could reach $100B
OpenAI’s fundraising conversations aren’t limited to Nvidia. CNBC has previously reported that the round could total around $100 billion, potentially closing in phases—starting with strategic investors and then widening to a larger pool.
If the numbers hold, the deal would stand as one of the largest private funding rounds ever attempted in tech—another signal that the AI arms race has shifted from model training alone to a capital-intensive battle over compute, infrastructure, and distribution.