Nvidia is putting as much as $105 billion behind the first phase of one of the world’s biggest AI data center projects, deepening its already massive relationship with OpenAI.
The deal gives us another clear sign of where the AI race is heading. The next phase of competition is no longer just about building better models. It is increasingly about who can secure enough chips, computing capacity, electricity and data center infrastructure to run those models at scale.
For Nvidia, it is a huge infrastructure commitment. For OpenAI, it helps secure access to the computing power needed to keep expanding. And for investors, it raises an important question: how much of the AI infrastructure boom is being driven by genuine end-user demand, and how much is being supported by financing arrangements between companies in the same ecosystem?
A massive data center deal takes shape
OpenAI has agreed to secure up to roughly 8 gigawatts of computing capacity from a planned data center campus in Pike County, Ohio.
The first 800 megawatts are expected to come online by 2028.
To put the scale into perspective, one gigawatt can provide enough power for as many as 750,000 US homes at a given time. The Ohio project would therefore be far larger than a conventional data center buildout.
The campus is expected to be equipped with Nvidia semiconductors and computing systems, making Nvidia both a supplier and a major financial backer of the infrastructure supporting its own hardware demand.
That is what makes this agreement particularly interesting.
Nvidia is backing the infrastructure, not the entire project
The headline figure is $105 billion, but there is an important distinction.
Nvidia’s commitment relates to the initial 4.25-gigawatt portion of the project. The company may later take on additional obligations connected to another 3.75 gigawatts.
The backing is expected to begin in phases between 2028 and 2030 and could remain in place for 20 years.
Nvidia has also made clear that it is not guaranteeing the financing of the entire site or all of OpenAI’s lease obligations.
OpenAI, meanwhile, will begin paying for the capacity as it becomes available for lease.
That structure matters because the economics of AI infrastructure are becoming increasingly complicated. Building these facilities requires enormous amounts of capital before the computing capacity can generate revenue.
The $600 billion opportunity for Nvidia
The Ohio project is only one piece of a much larger relationship.
Nvidia said OpenAI’s overall purchases of its equipment could represent around $600 billion in revenue through 2030.
That number highlights just how dependent the AI boom has become on Nvidia’s chips and systems.
As AI models grow larger and companies add more AI products, the amount of computing power required continues to increase. That creates an enormous opportunity for Nvidia, which sits at the center of the hardware supply chain.
The company is no longer simply selling chips into the AI market.
It is increasingly helping build the infrastructure that allows customers to buy and use those chips at scale.
But investors are watching the financing structure closely
There is another side to the story.
Investors have become increasingly focused on what are sometimes described as circular AI financing arrangements.
The concern is straightforward. If a hardware supplier provides financing or investment to an AI company, and that company then uses the money to purchase hardware from the same supplier, the resulting demand can look stronger than it might otherwise be.
Nvidia CEO Jensen Huang has pushed back on that argument.
He said the OpenAI agreement does not represent circular financing, arguing that Nvidia is supporting capacity when it has visibility into customer demand and when doing so helps create long-term productive infrastructure.
OpenAI will also be responsible for paying the lease as the capacity becomes available.
Still, the issue is unlikely to disappear.
The AI infrastructure market is moving at an extraordinary pace, with chipmakers, cloud providers, data center operators, energy companies and AI labs increasingly tied together through investments, supply agreements and financing commitments.
For investors, understanding those relationships may become just as important as looking at headline revenue growth.
SoftBank is also part of the picture
The project has another major player: SoftBank-backed SB Energy.
Nvidia has separately agreed to invest $1.5 billion in SB Energy, which will build, own and operate the facility.
That adds another layer to the ecosystem around the Ohio project.
SoftBank founder Masayoshi Son has been one of the biggest proponents of large-scale AI infrastructure investment, while the Trump administration has also highlighted the Ohio development as a major investment project.
The scale of the proposed campus makes it a symbol of the broader infrastructure race taking place around AI.
This is bigger than Nvidia and OpenAI
The Nvidia and OpenAI agreement shows how the AI industry is evolving.
A few years ago, much of the attention was focused on who could build the most capable AI model.
Today, the competitive advantage increasingly depends on infrastructure.
Companies need:
- Advanced chips to train and run AI models
- Massive data centers to house those systems
- Reliable electricity to keep them running
- Long-term financing to build the infrastructure
- Customers willing to commit to capacity for years into the future
That is why deals like the Ohio project matter.
They show that AI’s next bottleneck may not simply be software. It may be the physical infrastructure required to support the industry’s growth.
Nvidia is becoming more than a chip company
Nvidia’s strategy is becoming increasingly interesting from an investment perspective.
The company has already provided financial backing to other players building AI infrastructure, including CoreWeave. It has also invested directly in OpenAI and Anthropic.
Now, with the Ohio project, Nvidia is taking on its largest arrangement of this type so far.
That raises the potential rewards, but it also increases the importance of execution.
If AI demand continues to grow at the pace expected by the industry, Nvidia could benefit from an enormous expansion in computing infrastructure.
But if demand expectations prove too optimistic, the capital intensity of these projects could become a bigger concern.
The bigger investor question
Nvidia shares have already delivered a huge run, gaining around 21% this year as of the report.
The market is therefore looking beyond whether Nvidia can sell more GPUs.
The bigger question is whether the enormous infrastructure being built today will generate enough economic value to justify the capital being committed.
The Ohio project is a major vote of confidence in the future of AI.
It is also a reminder that the AI boom is becoming a highly interconnected financial ecosystem.
Nvidia needs OpenAI as a massive customer. OpenAI needs Nvidia’s computing infrastructure. Data center developers need long-term customers. And all of them need enormous amounts of capital and power.
That creates opportunities across the ecosystem, but it also means investors need to look beyond individual companies and understand how the pieces fit together.
The bottom line
Nvidia’s potential $105 billion commitment to the first phase of OpenAI’s Ohio data center project is one of the clearest examples yet of how large the AI infrastructure race has become.
The project could deliver enormous computing capacity to OpenAI while creating a potentially massive long-term revenue opportunity for Nvidia.
But the deal also puts the spotlight on the financing structures supporting the AI boom.
For investors, the story is no longer simply about who makes the best AI model or who sells the most chips.
It is about whether the entire infrastructure ecosystem being built around AI can generate the returns needed to support its extraordinary scale.
And that may be one of the most important investment questions of the next few years.