Nvidia to back Ohio AI campus plan for OpenAI, commits $1.5 billion to SB Energy

Nvidia has agreed to support a new, power-hungry artificial intelligence computing campus planned in Ohio that is intended to serve OpenAI, according to a report by Yahoo Finance published on Tuesday. The company’s backing centers on securing the infrastructure needed for what is described as an 8-gigawatt (8-GW) AI campus, a scale that underscores how electricity and grid capacity have become bottlenecks for large-scale AI training and inference.

Alongside the Ohio campus plan, Nvidia also said it will invest $1.5 billion in SB Energy, an energy developer. SB Energy is described in the report as the partner tied to building out the power and infrastructure required for major AI and data center projects. The combination of a technology supplier role and an energy investment highlights how chipmakers are increasingly moving beyond semiconductors to ensure the downstream systems that run their chips can be built and powered.

The Yahoo Finance report frames Nvidia’s Ohio involvement as a commitment to “secure infrastructure” for the campus rather than as a direct construction contract for the full data center complex. It does not, in the information provided for this story, specify the mechanism of support, such as ownership stakes, offtake arrangements, equipment supply terms, or the timeline for milestones. It also does not provide project-level details like the number of data centers, expected launch phases, or how the capacity will be allocated between different customers.

The agreement is notable for the scale of the campus target. An 8-GW facility would require significant electrical generation and transmission capacity, plus large on-site systems for power delivery, cooling, and redundancy. In practice, the ability to secure interconnection and long-lead electrical equipment often determines whether large AI deployments can move from planning to operation. For Nvidia, that matters because the company’s data center GPUs and related software typically need ample, continuously available compute to deliver the throughput AI customers are seeking.

For OpenAI, the report positions the Ohio campus as part of a broader effort to expand AI compute capacity. While the details of OpenAI’s usage plan are not included in the information provided here, the overall structure of such campus projects generally aims to provide dedicated or prioritized compute for training and large-scale deployment workloads, where delays can cascade into product timelines and model iteration schedules.

Nvidia’s $1.5 billion investment in SB Energy adds to a growing theme across the AI supply chain: power is increasingly treated as a strategic asset. Developers with experience in permitting, grid negotiations, and power delivery are often better positioned to execute than individual technology firms acting alone. By investing directly in an energy developer, Nvidia may seek to reduce execution risk for future deployments that depend on reliable power and rapid expansion.

Still, important questions remain unanswered based on the disclosed information. The report does not provide the expected start of construction, the expected completion dates, or whether the campus will be reserved exclusively for OpenAI or shared across multiple customers. It also does not disclose how much of the 8-GW target is already under contract, whether Nvidia’s infrastructure support includes specific power capacity commitments, or how the economics of the arrangement are structured.

Investors and industry watchers will likely focus next on clarifications around the Ohio campus execution, including permitting progress, grid interconnection milestones, and how the project fits into Nvidia and OpenAI’s near-term AI capacity plans. The deal’s follow-through may also be read as a announcement of how seriously Nvidia views power infrastructure constraints as a competitive and operational variable, not just a background cost.