Sources: OpenAI is in advanced talks to lease a proposed 10GW, $500B+ data center campus in Ohio as part of a deal that could include Nvidia's financial backing
OpenAI is in advanced negotiations to lease a proposed 10 gigawatt data center campus on federal land in Ohio as part of a deal …
Context & Ripple Effects
OpenAI’s infrastructure push has moved from plans for Midwest and Southwest facilities to an expanded Oracle arrangement targeting 5GW, alongside scouting for a 1GW-plus Indian site. The Ohio discussion would be a much larger concentration of capacity than those related reports describe.
Nvidia has already been linked to OpenAI through proposed 10GW system deployments, potential investment, and discussions of chip leasing rather than outright purchases. The reported Ohio structure would extend that relationship from hardware supply toward financing and long-lived infrastructure commitments.
First-order effects
- OpenAI could secure control of a proposed 10GW Ohio campus through a lease, reducing the need to directly own the site while committing it to a very large single-campus capacity buildout.
- Nvidia could become both a capital partner and infrastructure supplier, deepening its commercial exposure to OpenAI beyond chip sales.
Second-order effects
- A lease-and-financing model could make it easier for OpenAI to reserve capacity before fully funding every component of construction, while tying future infrastructure choices more closely to Nvidia’s platform.
- Other hyperscale AI builders and data-center partners may face sharper competition for power-ready sites and for financing structures that pair compute vendors with major model developers.
Third-order effects
- If repeated, these arrangements would shift AI infrastructure from conventional equipment procurement toward vertically linked capital, chip, and capacity partnerships between model labs and their key suppliers.
- The scale of proposed campuses would make access to power, land, and permitting an increasingly decisive constraint on AI expansion, rather than accelerator availability alone.
The trend: AI labs are evolving into infrastructure-scale buyers whose compute expansion is financed and structured through increasingly strategic partnerships with chip vendors and data-center operators.