Open AI3 mins read

OpenAI’s Ohio Data Center Lease: What the Nvidia-Backed Deal Means

OpenAI signed a 20-year lease for a massive Ohio AI data center project, with Nvidia backing up to $105 billion and becoming the exclusive chip supplier for the first half of the site.

OpenAI and SB Energy logo lockup for the PORTS-Pike data center project
Image credits:Nvidia

The Core Deal: 20 Years, Ohio, and 8 Gigawatts

OpenAI has signed a 20-year lease with SoftBank subsidiary SB Energy for the PORTS-Pike campus in Ohio. The deal gives OpenAI around 8 gigawatts of IT capacity, while the Wall Street Journal puts the project’s gross total at 10 gigawatts once cooling and infrastructure are included. The site is described as partly located on a former US Department of Energy uranium enrichment facility and tied to power from a 9.2-gigawatt gas plant.

Nvidia’s Role Goes Beyond Supplying Chips

Nvidia is backing the project with up to $105 billion, but the arrangement is not a guarantee of OpenAI’s rent payments. According to the article, Nvidia is backing the residual value of finished data centers in the first construction phase, which covers 4.25 gigawatts of IT capacity. If OpenAI exits, SB Energy would first need to seek another tenant and then try to sell the facilities before Nvidia covers any remaining value difference, up to the cap. In return, Nvidia becomes the exclusive chip supplier for the first half of the site and is investing $1.5 billion in SB Energy.

Why ‘Land, Power, and Shell’ Now Matter

Nvidia CEO Jensen Huang framed the bottleneck with the acronym “LPS,” meaning land, power, and shell. The article says these basics have become a strategic constraint as AI labs expand faster than their balance sheets can support long-term infrastructure contracts. For readers tracking AI competition, the takeaway is clear: compute growth is now tied as much to physical infrastructure and financing structures as to chips themselves.

The Bigger Risk: AI Commitments Off the Balance Sheet

The Wall Street Journal analysis cited by The Decoder says nine tech companies hold around $3 trillion in mostly AI-related obligations that do not appear on their balance sheets. The article notes that leases are recorded only once payments begin, and purchase commitments only when goods are delivered. Analysts cited in the report warn that investors can barely gauge actual debt levels, making AI infrastructure exposure harder to evaluate from standard financial statements alone.

What to Watch Next

OpenAI says it only pays for finished capacity, and the first 800 megawatts are slated to come online in 2028. Huang expects OpenAI’s commitments through 2030 to reach roughly 12 gigawatts of Nvidia compute across all sites. If Nvidia exercises its option on the remaining 3.75 gigawatts in Ohio, the package could grow to about 16 gigawatts worth roughly $600 billion, according to the article.

Discover More