Nvidia Could Pour $3 Billion Into SoftBank’s SB Energy. The Bigger OpenAI Deal Could Reach $100 Billion.

Nvidia is reportedly considering investing as much as $3 billion in SoftBank-backed SB Energy as part of a much larger financing package for an OpenAI data center project in Ohio, highlighting the enormous sums flowing into infrastructure needed to power artificial intelligence.

Nvidia, OpenAI and SB Energy are also discussing around $100 billion in credit support for the planned Ohio campus, which would require massive amounts of computing hardware, electricity and data center capacity.

The discussions could further deepen the financial ties between some of the biggest beneficiaries of the AI boom, with Nvidia potentially providing both the chips and financial backing needed to build infrastructure used by OpenAI.

The Information noted that Nvidia has discussed investing half of the proposed $3 billion when the Ohio project agreement is signed and the remaining half as part of a planned SB Energy initial public offering.

The deal has not been finalized, and Nvidia and SB Energy did not immediately comment on the report.

SB Energy is also reportedly preparing to go public as soon as next month and could seek to raise at least $5 billion through the offering.

The company, founded in 2019 and backed by SoftBank and OpenAI, develops large-scale energy and data center infrastructure. Its role has become increasingly important as AI companies search for ways to secure the enormous quantities of power and computing capacity required for their expansion.

The potential Nvidia investment comes as the chipmaker reworks its involvement in the Ohio project.

The Wall Street Journal reported that Nvidia had scaled back an earlier proposal to guarantee as much as $250 billion in financing for the OpenAI data center project. The company is now expected to initially guarantee less than $120 billion.

The shift illustrates both the extraordinary size of the financing being discussed and the financial risks involved in the AI industry's infrastructure race.

AI companies are committing hundreds of billions of dollars to new data centers as increasingly sophisticated models demand more advanced chips, electricity and cooling capacity. Nvidia sits at the center of that buildout because its graphics processors remain critical to training and running many of the world's most powerful AI systems.

But the company is increasingly doing more than simply selling chips.

Nvidia has invested across the AI ecosystem, backing model developers, cloud infrastructure companies and other businesses whose expansion can ultimately increase demand for its hardware.

Its relationship with OpenAI has become particularly significant.

Nvidia and OpenAI announced a strategic partnership in September 2025 under which OpenAI planned to deploy at least 10 gigawatts of Nvidia systems for its next-generation AI infrastructure. Nvidia said at the time that it intended to invest as much as $100 billion in OpenAI as the infrastructure was deployed.

The latest discussions involving SB Energy would potentially expand that relationship further by bringing Nvidia into the financing of the physical infrastructure needed to house and power those systems.

OpenAI, meanwhile, has been aggressively expanding its data center ambitions through its Stargate initiative.

OpenAI said last year that it was working with Oracle and SoftBank on five additional U.S. data center sites, putting Stargate on a path toward nearly 7 gigawatts of planned capacity and more than $400 billion in investment over three years.

Ohio has emerged as an important piece of those ambitions.

SB Energy and OpenAI have been developing large-scale data center infrastructure in the state, while the energy company has also been expanding projects designed to supply the huge electricity requirements created by AI workloads.

The financing arrangements demonstrate one of the biggest challenges facing the AI industry: finding enough capital to build infrastructure at the speed companies such as OpenAI want to expand.

The largest AI models require clusters containing enormous numbers of expensive processors. Those chips must then be housed in specialized data centers connected to sufficient power generation and transmission capacity.

That means the AI race is increasingly becoming an infrastructure and financing race as much as a competition to build better models.

Nvidia has a clear incentive to help that infrastructure get built. More AI data centers potentially mean more demand for Nvidia chips, networking equipment and other systems.

But the growing financial connections between Nvidia, AI developers and infrastructure providers have also drawn attention to how money circulates through the industry. Nvidia can invest in companies that subsequently use financing to purchase large quantities of Nvidia hardware, creating increasingly intertwined relationships across the AI ecosystem.

The proposed SB Energy investment would be another example of that trend.