Nvidia’s $105 Billion OpenAI Bet: Why the Massive Ohio AI Data Center Deal Matters

The Nvidia OpenAI data center deal in Ohio is one of the clearest signs yet that the artificial intelligence boom is moving beyond chips and software into land, electricity, financing and industrial-scale construction. Nvidia has agreed to provide up to $105 billion in guarantees connected to OpenAI’s long-term lease at the PORTS-Pike Technology Campus in Pike County, Ohio, while also investing $1.5 billion in developer SB Energy. The project is designed to provide OpenAI with roughly 8 gigawatts of IT capacity over time.

The headline number is enormous, but it needs context. Nvidia is not simply handing OpenAI a $105 billion cheque. The guarantee is designed to support part of the lease and power obligations and to help protect a minimum value for the site if OpenAI cannot meet its commitments. That structure can make it easier for SB Energy to finance construction while giving Nvidia a direct role in securing future demand for its own computing systems.

The Nvidia OpenAI data center therefore matters for more than OpenAI. It shows how closely the AI industry’s biggest companies are becoming tied together through chips, capital, energy and long-term infrastructure commitments. It also raises a harder question: can demand for AI services grow fast enough to justify projects that require tens of billions of dollars, years of construction and power generation measured in gigawatts?

Nvidia OpenAI Data Center Deal at a Glance

Detail	What has been announced
Location	PORTS-Pike Technology Campus, Pike County, Ohio
OpenAI capacity	Approximately 8 IT-GW
Initial Nvidia-backed capacity	4.25 IT-GW
Nvidia guarantee	Up to $105 billion
Nvidia investment in SB Energy	$1.5 billion
OpenAI lease	20 years
First capacity expected	2028
New power generation planned	At least 10 GW
Regional grid investment	At least $4.2 billion
Construction jobs projected	About 35,000 through 2032
Long-term operating jobs	About 2,500

The Nvidia OpenAI data center is being developed by SB Energy, which will build, own and operate the facility. Nvidia will be the exclusive AI compute infrastructure provider for the initial deployment, while OpenAI will use the capacity as the customer. The initial 4.25 IT-GW deployment can later expand by another 3.75 IT-GW.

1. The $105 Billion Figure Is a Guarantee, Not a Cash Investment

The most important detail in the Nvidia OpenAI data center story is also the easiest to misunderstand.

Nvidia has agreed to provide a guarantee of up to $105 billion, but that does not mean the chip company is immediately spending $105 billion on construction. The guarantee supports a portion of the lease and power obligations associated with the site and includes protection intended to ensure the completed infrastructure retains a minimum value.

If OpenAI were unable to meet its obligations, Nvidia could be responsible for covering part of the gap between that guaranteed value and what the owner could recover by finding another tenant or selling the asset. That makes the arrangement a significant financial commitment, but it is different from an ordinary equity investment or direct cash purchase.

Separately, Nvidia is putting $1.5 billion directly into SB Energy. That investment joins capital already committed by SoftBank and OpenAI and gives Nvidia a closer financial relationship with the company responsible for developing the campus.

For SB Energy, Nvidia’s backing can make an enormous construction project easier to finance. For Nvidia, the logic is strategic. The Nvidia OpenAI data center would be filled with Nvidia hardware, giving the chipmaker a path to years of future sales while securing a site where successive generations of AI systems can be installed.

This is why the guarantee should be viewed as both financial support and demand creation.

2. Nvidia Is Trying to Secure Demand for Generations of AI Chips

Nvidia’s traditional advantage has been selling the hardware used to train and run advanced AI models. The Nvidia OpenAI data center extends that strategy further into the infrastructure around those chips.

SB Energy says the campus will use Nvidia’s DSX AI factory platform, including GPUs, CPUs and networking. Nvidia will also be the exclusive AI compute infrastructure provider at PORTS-Pike for the initial capacity.

That matters because the life of a data center is much longer than the life of an individual chip generation. A campus can remain operational for decades while servers are replaced repeatedly. Nvidia CEO Jensen Huang said the company wants to secure long-lived infrastructure that can be upgraded as new generations of computing arrive.

For Nvidia, that could convert a one-time hardware sale into a recurring cycle of upgrades.

Reuters reported that Huang said the initial 4.25-gigawatt portion of the Nvidia OpenAI data center could eventually contribute as much as $200 billion to Nvidia revenue. He also said Nvidia could make as much as $600 billion from OpenAI by 2030 across 16 gigawatts of computing capacity. These are Nvidia projections, not guaranteed future revenue.

The News Ink has already examined how rapidly Nvidia’s AI business has expanded in our coverage of the company’s $81.6 billion earnings report. The Ohio project suggests that Nvidia increasingly sees access to land and electricity as nearly as important to future hardware sales as the chips themselves.

3. OpenAI Needs Computing Power on an Industrial Scale

The Nvidia OpenAI data center is also a window into the infrastructure demands of frontier AI.

OpenAI says it has agreed to secure approximately 8 gigawatts of IT capacity at PORTS-Pike. Reuters noted that one gigawatt of power is roughly comparable to the average electricity consumption of about 750,000 U.S. homes, which helps illustrate the extraordinary scale involved.

This does not mean the Nvidia OpenAI data center will literally consume the same electricity as six million homes at every moment. IT capacity, generating capacity and electricity consumption are not identical measures. The comparison simply puts the size of the proposed infrastructure into perspective.

OpenAI is effectively planning infrastructure at a scale historically associated with major industrial facilities rather than conventional software companies.

That is one reason the Nvidia OpenAI data center is being developed around the former Portsmouth Gaseous Diffusion Plant in Pike County. The U.S. Department of Energy is leasing federal land at the former uranium-enrichment site for the development, part of a wider effort to reuse the industrial property.

OpenAI says capacity will arrive in phases beginning in 2028, while its projected six-year construction period extends through 2032.

The infrastructure challenge is not unique to Ohio. The News Ink previously reported on OpenAI pausing a UK data-centre project, highlighting how energy availability and economics can become decisive constraints as AI companies pursue larger computing clusters.

4. Electricity Has Become One of AI’s Biggest Bottlenecks

The Nvidia OpenAI data center would not be possible without a parallel energy buildout.

SB Energy and SoftBank plan to develop at least 10 gigawatts of new power generation to support the campus. The U.S. Department of Energy says the wider plan includes at least 9.2 GW of natural-gas generation, while SB Energy and AEP Ohio are planning at least $4.2 billion in new transmission infrastructure.

The national trend is becoming increasingly visible.

The U.S. Energy Information Administration estimates that data-center servers accounted for about 7% of U.S. commercial-sector electricity consumption in 2025. Across its 2026 long-term scenarios, server electricity use could grow to between 22% and 33% of commercial-building electricity consumption by 2050.

The Nvidia OpenAI data center brings that wider trend into a single project of exceptional scale.

The Department of Energy says SB Energy has committed to paying for the new transmission infrastructure associated with the campus instead of shifting those construction costs directly onto ordinary regional ratepayers. OpenAI similarly says the project will cover its project-specific energy and infrastructure costs.

Those commitments will be closely watched.

Large data centers have increasingly generated debate over electricity availability, grid expansion, water consumption and the effect of rapid demand growth on surrounding communities.

OpenAI says the Ohio facility will use closed-loop, air-cooled cooling systems that recirculate water instead of conventional cooling towers that continuously consume water. The company has pledged to publish future information on expected water use once the design is finalized and to report project-related water and energy use annually.

5. The Deal Shows Why AI Financing Is Becoming More Complicated

The Nvidia OpenAI data center is financially unusual because the supplier, customer, developer and investors are increasingly interconnected.

Nvidia sells the computing hardware. OpenAI needs huge quantities of computing capacity. SB Energy builds the site. SoftBank backs SB Energy and also has strategic interests in AI infrastructure. Nvidia is investing in SB Energy while providing financial support connected to infrastructure where Nvidia equipment will run.

That structure has intensified debate about circular financing in the AI industry.

Reuters noted that critics have questioned arrangements in which companies financing AI infrastructure can also benefit from the resulting purchases of their own products. Huang rejected the suggestion that the Nvidia OpenAI data center constitutes circular financing, arguing that Nvidia is using its scale and visibility into demand to secure scarce infrastructure.

The distinction matters.

Circular relationships do not automatically mean that a transaction is economically unsound. Suppliers have financed customers in other capital-intensive industries, while long-term tenant agreements routinely help developers obtain funding for large projects.

The real test is whether the underlying business eventually produces enough revenue and cash flow to justify the money committed.

That question has become particularly important as valuations of frontier AI companies surge. The News Ink recently examined the enormous expectations being placed on the industry in our analysis of the Anthropic IPO.

The Nvidia OpenAI data center represents the physical side of the same bet. Investors are assuming that future demand for AI will be large enough to support today’s extraordinary infrastructure spending.

Earlier reporting had suggested Nvidia might provide guarantees worth as much as $250 billion around the Ohio project, but the structure ultimately announced on August 17 was considerably narrower, with the guarantee capped at up to $105 billion.

That change is significant because it shows the companies are trying to balance the need for enormous financing with investor concerns about Nvidia taking excessive exposure to its customers.

6. Ohio Could Gain Jobs and Investment, but the Local Impact Matters

OpenAI says the Nvidia OpenAI data center could create about 35,000 construction jobs during the six-year buildout through 2032 and approximately 2,500 long-term operating jobs.

The companies are also promising substantial community investment.

SB Energy previously established a $40 million community benefits fund. OpenAI has committed another $40 million, bringing the planned fund to $80 million. OpenAI also says it will provide $84 million in Codex credits through ChatGPT for Ohio college students.

OpenAI says the community funding could support local priorities including schools, public safety, healthcare, utilities, workforce training, housing, veterans’ services and small businesses. The company also expects the project to generate significant state and local tax revenue over its operating life, although the eventual amount will depend on the project actually being built and operated as planned.

The location carries symbolic importance too.

The PORTS-Pike campus is being created around the former Portsmouth Gaseous Diffusion Plant, which played a role in U.S. uranium enrichment during the Cold War. Federal officials now see the redevelopment as part of a broader reindustrialization effort in Southern Ohio.

Supporters therefore see the Nvidia OpenAI data center as a chance to turn an older industrial site into infrastructure for a new technological industry.

OpenAI says it will publish annual reports covering local hiring, community investment, water use and project-related energy use. Those reports should provide a useful benchmark for comparing the benefits promised now with the results experienced by Pike County later.

7. The Biggest Risk Is Whether AI Revenue Can Catch Up With AI Spending

The central question behind the Nvidia OpenAI data center is simple: will the economics eventually work?

The project assumes demand for AI services will continue expanding for years. OpenAI must generate enough revenue to support enormous long-term infrastructure commitments. Nvidia needs sustained demand for its hardware. SB Energy needs the campus to remain valuable. Energy companies need major new generation and transmission projects to make economic sense.

If AI adoption continues accelerating, the Ohio campus could eventually resemble an early investment in infrastructure that became essential.

But the opposite scenario also deserves attention.

If models become dramatically more efficient, if competition pushes AI prices lower, if businesses prove unwilling to pay for premium services or if new computing architectures require substantially less power, some of today’s planned capacity could eventually prove excessive.

There is also technological risk.

Data centers are long-lived assets, while AI hardware evolves extremely quickly. A facility planned today needs to remain adaptable enough to accommodate computing systems that may look very different in five, 10 or 20 years.

Nvidia argues that the Nvidia OpenAI data center will be designed so its computing systems can be upgraded repeatedly as new generations arrive. That reduces dependence on one particular GPU generation, but every upgrade still requires capital, electricity, cooling and customers willing to pay for the resulting AI capacity.

This tension is appearing across the industry.

OpenAI, Anthropic, cloud providers and chipmakers are competing for advanced semiconductors, electricity, transmission capacity, skilled workers and suitable land.

The companies that secure those resources efficiently could build durable advantages. Those that overbuild could eventually face weak returns or costly write-downs. That makes the Nvidia OpenAI data center a test of both technological demand and capital discipline.

Why the Deal Matters Beyond Nvidia and OpenAI

The Nvidia OpenAI data center is not simply another corporate partnership. It reveals how the structure of the AI industry is changing.

Compute capacity is becoming strategic infrastructure. Nvidia is moving beyond being only a component supplier and is helping secure the physical sites where future generations of its hardware can operate.

OpenAI, meanwhile, is securing long-term capacity on a scale more commonly associated with factories, energy projects and other industrial megaprojects.

Government is also part of the equation. The Department of Energy is leasing federal land at PORTS, while new energy generation and transmission infrastructure are being developed alongside the computing campus.

The Nvidia OpenAI data center also shows that compute, energy, financing and government policy are becoming part of one AI infrastructure market. It could become a model for how frontier AI capacity is financed and built.

That makes the project relevant not only to Nvidia and OpenAI shareholders, but also to utilities, construction companies, energy producers, communities, regulators and anyone trying to understand where the AI boom goes next.

What to Watch Next

Several milestones will determine whether the project develops as planned.

Financing: The $105 billion guarantee can support development, but building the full campus still requires enormous amounts of capital.

Construction: Initial capacity is expected beginning in 2028, with the broader buildout continuing through 2032.

Power generation: At least 10 GW of new generation and $4.2 billion of transmission investment have to arrive alongside the computing infrastructure.

OpenAI growth: The economics ultimately depend on demand for OpenAI’s services continuing to expand.

Nvidia hardware cycles: The value of the Nvidia OpenAI data center partly depends on the campus being able to replace older systems with newer Nvidia computing platforms.

Community impact: Local employment, taxes, housing, infrastructure, water consumption and energy prices will determine how residents ultimately judge the development.

AI efficiency: If future models can perform far more work using much less computing power, long-term infrastructure requirements could differ substantially from current forecasts.

Frequently Asked Questions

Is Nvidia investing $105 billion directly in the Ohio data center?

No. The Nvidia OpenAI data center arrangement includes a guarantee of up to $105 billion connected to portions of the lease, power obligations and minimum infrastructure value. Nvidia is separately investing $1.5 billion directly in SB Energy.

How large will the Ohio AI data center be?

OpenAI says it has agreed to secure approximately 8 IT-gigawatts at PORTS-Pike. The initial Nvidia-backed deployment is designed for 4.25 IT-GW, with Nvidia holding an option connected to another 3.75 IT-GW.

When will the data center start operating?

The first capacity at the Nvidia OpenAI data center is expected to come online in phases beginning in 2028. OpenAI says the wider construction program is expected to run through 2032.

Why is Nvidia backing infrastructure for OpenAI?

Nvidia says securing long-term land, power and building capacity helps ensure there are locations where future generations of Nvidia computing systems can be deployed. The arrangement also supports a major customer expected to require enormous amounts of AI infrastructure.

Will the project increase electricity prices in Ohio?

The developers and U.S. Department of Energy say SB Energy will pay the project-specific costs of transmission and energy infrastructure rather than shifting those costs directly to ordinary regional ratepayers. The project’s actual effect on the wider electricity system will be something to monitor as construction progresses.

Conclusion

The Nvidia OpenAI data center deal is one of the strongest examples yet of how rapidly artificial intelligence is becoming a physical infrastructure industry.

Nvidia’s commitment of up to $105 billion is not a simple investment cheque. It is a financial guarantee designed to help secure land, power and long-lived data-center infrastructure for OpenAI, while Nvidia separately invests $1.5 billion in SB Energy and positions its technology at the centre of the Ohio campus.

For OpenAI, the Nvidia OpenAI data center offers access to roughly 8 gigawatts of future IT capacity. For Nvidia, it could lock in years of demand for GPUs, CPUs, networking and future hardware generations. For Southern Ohio, the development promises tens of thousands of construction jobs, major grid investment and a new industrial role for a former uranium-enrichment site.

But the scale is also the risk.

The Nvidia OpenAI data center assumes AI demand, revenues and infrastructure needs will remain enormous well into the next decade. It assumes OpenAI can support long-term obligations, that Nvidia systems will remain central to advanced AI and that massive investments in electricity generation and transmission can be justified economically.

If those assumptions hold, PORTS-Pike could become one of the defining infrastructure projects of the AI era. If they do not, the $105 billion guarantee could instead become a symbol of how aggressively the industry financed expectations of future AI demand.

Either way, the Nvidia OpenAI data center makes one thing clear: the AI race is no longer only about who builds the best model. It is increasingly about who can secure the land, electricity, capital and computing capacity required to run those models at unprecedented scale.

Follow The News Ink

Stay connected with The News Ink for the latest technology, artificial intelligence, world news, science and sports coverage.

Follow us on X, Instagram, Threads, TikTok, Bluesky and Mastodon.

For longer-form coverage and discussion, follow The News Ink on Medium, Substack and Quora, or discover visual stories on Pinterest.