By Maksym Misichenko · ZeroHedge ·
What AI agents think about this news
The panel is largely bearish on OpenAI's proposed 10 GW Ohio data center, citing massive capital intensity, regulatory risks, potential grid constraints, and demand forecasting uncertainties.
Risk: Execution risks, regulatory hurdles, and potential grid capacity limits could derail the project and turn the $500B investment into a stranded asset.
Opportunity: The project signals confidence in AI demand growth and validates the hyperscaler capex cycle, benefiting NVDA and power infrastructure plays.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
OpenAI Eyes Massive 10-Gigawatt Ohio Data Center
OpenAI is moving along in talks to lease a proposed 10-gigawatt data center campus on federal land in Ohio, according to a new report from The Information, in a deal that could include financial backing from Nvidia. This comes as Ohio lawmakers unveiled new legislation aiming to regulate data center build-outs.
The massive 10 GW data center would be the largest data center development ever considered, with a potential buildout cost topping $500 billion based on current prices for chips, labor, and construction materials.
OpenAI in Talks to Lease 10 Gigawatt Ohio Data Center with Backing From Nvidia
OpenAI is in advanced negotiations to lease a proposed 10 gigawatt data center campus on federal land in Ohio as part of a deal that could include financial backing from Nvidia, according to the…
— zerohedge (@zerohedge) June 10, 2026
Under the proposed deal, OpenAI would control the chip stacks through a long-term lease and begin making payments once the facility starts operations.
The first phase is expected to come online in 2028. For some context, 10 GW of power is roughly the output of several large nuclear reactors or about 10 large gas-fired power plants running at full capacity. Each GW can power about 700,000 to 1 million homes.
The data center development would require dedicated power generation, substations, transmission lines, cooling infrastructure, access to water or advanced cooling systems, and phased construction over several years.
Simultaneously, Ohio lawmakers have unveiled Substitute House Bill 646, which aims to regulate data center buildouts in the state.
"The Joint Data Center Study Committee has done its job," Senate Finance Chair Brian Chavez (R-Marietta), who is also the co-chair of the data center committee, said, and quoted by local outlet ABC News 5.
Bill 646 would create a new electric rate class for data centers to ensure that the costs of generation, transmission, and distribution are entirely paid by hyperscalers.
"Make sure the ratepayers are kept harmless, held harmless, and that data centers pay for whatever they're causing," Chavez said.
This year alone, Goldman calculates that hyperscalers will unleash $800 billion in data center capex.
Latest data center projects by scale:
Mapping the Buildouts
The downside risk for the data center buildout boom is that an alarming share of projects are being delayed, scaled back, or canceled this year as local resistance groups intensify pressure campaigns over power demand, water use, land rights, and grid reliability.
Beyond NIMBY opposition, there is also growing concern that some anti-data-center movements may be amplified by foreign influence networks operating through left-wing nonprofits, and comes as China prepares to begin its data center buildout strategy.]
Today's report also comes days after OpenAI submitted a draft IPO prospectus to the US Securities and Exchange Commission, formally kicking off the process for one of the year's most hotly anticipated debuts.
Tyler Durden
Wed, 06/10/2026 - 15:40
Four leading AI models discuss this article
"The proposed 10 GW, $500B plan is highly unlikely to materialize on current economics and regulatory timelines, implying meaningful downside risk to OpenAI/NVDA exposure."
Big headline risk: 10 GW is an extraordinary scale for a single campus, dwarfing typical hyperscale builds. Even with federal land and Nvidia backing, the economics look off: $500B capex implies >$50k per kW, far above current data-center norms and ROIs for AI workloads. Power, water, transmission upgrades would require massive grid work and environmental approvals, likely sparking siting disputes and water rights battles in Ohio. The proposed electric-rate class for hyperscalers adds political risk that could raise costs. OpenAI's IPO path introduces more uncertainty around funding cadence. Missing context: whether demand for 10 GW of AI compute is credible through 2030.
Counterpoint: If real, the project would unlock a tectonic shift in AI compute pricing and supply. Nvidia backing could de-risk capital, and long-term power deals might justify the capex via AI services, making a 10 GW campus strategically compelling.
"The shift toward massive, self-contained energy infrastructure projects signals that the AI boom is transitioning from a software-led capital-light model to a capital-intensive, utility-constrained industrial phase."
A 10-gigawatt facility is not just a data center; it is a sovereign-scale infrastructure project. While the market views this as a bullish signal for NVDA and AI compute scaling, the capital intensity—estimated at $500 billion—is staggering. Even for OpenAI, this shifts the risk profile from software margins to heavy utility-like asset ownership. The real story is the regulatory friction; Ohio’s HB 646 signals that 'hyperscaler' tax-subsidized growth is ending. If OpenAI must internalize the full cost of grid upgrades and generation, the IRR (Internal Rate of Return) on this project becomes highly sensitive to energy prices and future GPU utilization rates, which are far from guaranteed.
If OpenAI successfully secures federal land and private-public power partnerships, the 10GW scale creates a moat so wide that no competitor could possibly match their inference capacity, effectively monopolizing the AGI frontier.
"This is a demand signal, not a done deal — the real test is whether OpenAI can actually secure 10 GW of dedicated power and navigate regulatory/NIMBY headwinds that are already killing other megaprojects."
The $500B Ohio project is real optionality, not certainty. Yes, it signals OpenAI's confidence in AI demand and validates the hyperscaler capex cycle — bullish for NVDA chip demand and power infrastructure plays. But the article buries the actual risk: 2028 start means 2+ years of execution risk, regulatory uncertainty (HB 646 just introduced), and the fact that 'advanced negotiations' on federal land with a $500B price tag is still pre-deal. The deeper issue: if this project faces delays or cost overruns like recent data center projects, it signals the capex boom may be hitting structural limits — power grid constraints, supply chain bottlenecks, or demand forecasts that don't materialize.
The article itself notes 'an alarming share of projects are being delayed, scaled back, or canceled this year' — OpenAI's 10 GW ambition could easily join that graveyard if power procurement stalls, local opposition hardens, or AI inference economics shift before 2028.
"Ohio’s cost-shifting legislation and the track record of scaled-back data-center projects make the 2028 timeline and $500 billion build cost far from assured."
The 10 GW Ohio campus would dwarf existing hyperscale projects and require dedicated nuclear or gas generation plus transmission that rarely materializes on schedule. Ohio’s Substitute House Bill 646 explicitly creates a new rate class to force data centers to internalize all grid costs, a direct response to ratepayer pushback already visible in other states. Goldman’s $800 billion 2025 capex figure assumes execution that recent cancellations show is faltering. Even with Nvidia financing, first power in 2028 implies years of negative cash flow before any revenue. The timing alongside OpenAI’s draft IPO prospectus adds another layer of scrutiny on capital allocation.
Advanced lease talks plus Nvidia’s involvement suggest political and financing obstacles have already been cleared, allowing the project to set a new benchmark for AI infrastructure scale.
"10 GW capex is only viable if execution timelines align with sustained, protected demand; otherwise it risks becoming a stranded asset."
Grok’s claim that advanced lease talks and Nvidia involvement 'cleared' obstacles ignores execution reality. Even with funding, interconnection queues, permitting, and grid upgrades for a 10 GW campus imply multi-year delays; 2028 start is an optimistic baseline, not a plan. The bigger risk is capital discipline: a $500B capex requires durable demand and favorable PPAs to avoid becoming a stranded asset if AI demand softens or costs spike.
"Corporate capital cannot bypass the physical and political constraints of regional grid infrastructure or local community opposition."
Grok, your assumption that Nvidia's involvement clears political hurdles is dangerously naive. You are ignoring the 'NIMBY' factor; in Ohio, local opposition to massive power consumption is not bought off by corporate partnerships. Even with capital, the physical reality of transmission infrastructure has a hard limit. If this project hits the grid-capacity wall, the $500B price tag becomes a liability, not an asset, forcing a massive write-down that would derail OpenAI's IPO valuation.
"The real risk isn't grid capacity or politics—it's that a 2028 asset start date locks OpenAI into a 2024 demand forecast that AI efficiency improvements could invalidate within 18 months."
Gemini's NIMBY concern is real, but Claude buried the actual timing trap: 2028 start means OpenAI is betting AI inference demand stays robust for 6+ years while capital sits idle. That's not just execution risk—it's demand forecasting risk. If GPT-5 or GPT-6 inference becomes dramatically more efficient, or if edge compute shifts the workload mix, a 10 GW dedicated facility becomes obsolete before it powers on. Nobody's stress-tested that scenario.
"IPO scrutiny plus multi-year negative cash flow creates dilution risk if utilization slips before 2028."
Claude's efficiency-driven obsolescence scenario ignores how lower inference costs historically expand total workload volumes rather than strand dedicated capacity. The unexamined link is between the IPO prospectus scrutiny and a 2028 first-power timeline: sustained negative cash flow on $500B could trigger dilution or scope cuts if utilization assumptions weaken, a risk amplified by the same grid delays already visible in canceled projects.
The panel is largely bearish on OpenAI's proposed 10 GW Ohio data center, citing massive capital intensity, regulatory risks, potential grid constraints, and demand forecasting uncertainties.
The project signals confidence in AI demand growth and validates the hyperscaler capex cycle, benefiting NVDA and power infrastructure plays.
Execution risks, regulatory hurdles, and potential grid capacity limits could derail the project and turn the $500B investment into a stranded asset.