Why Ohio

Ohio is not the obvious answer if you are reading old maps of the internet. It becomes the obvious answer the moment you start from the real bottleneck. The constraint in AI infrastructure is not demand, not capital, and not even chips by themselves. The constraint is buildable, powered, politically acceptable land. Filter the country through power, gas, fiber, water, workforce, permitting, and community posture — instead of legacy cloud geography — and Ohio stops looking like a secondary market and starts looking like a first-principles development state.

Start with energy. Ohio sits inside one of the most consequential energy corridors in the country: the U.S. Energy Information Administration puts it among the top ten natural-gas states, at roughly 5% of both national production and reserves. That matters because AI loads are now chasing firm, dispatchable power as hard as they are chasing fiber. It does not mean every campus should be gas-fired — it means every serious campus needs a realistic power plan, because the grid alone cannot absorb unlimited new load on fantasy timelines, as the DOE/LBNL demand curve makes plain. Ohio is already in the middle of the shift: the Office of the Ohio Consumers' Counsel counts more than 200 data centers in the state — fifth-most in the nation — with up to $40 billion in additional investment planned by 2030.

But Ohio's opportunity will be wasted if it runs the old playbook. That playbook is land-first: find acreage, promise jobs, apply for abatements, push the utility to solve power, tell the town it is lucky, and fight the backlash later. It is breaking everywhere. NCSL reports that roughly 14 states are weighing data-center bans or moratoriums, and more than 300 data-center bills were filed across 30 states in the first six weeks of 2026 alone. Ohio itself now has an active petition drive to put a data-center ban on the ballot, and the governor has paused data-center tax breaks after they cost far more than expected. Reuters/Ipsos finds 57% of Americans would oppose a data center nearby. The answer for Ohio is not to say yes to everything — it is to build the right template before the wrong one poisons the market.

The screening order, inverted

Powered land comes first, and powered land is not land with a sales brochure. It is land where load can actually be served: gas availability, grid redundancy, a transmission pathway, substation feasibility, honest interconnection math, and a load profile a utility or behind-the-meter developer can underwrite. Water comes next — data centers and greenhouses both need serious water planning, and the move is to design a campus water budget (withdrawals, reuse, stormwater, wastewater, contingencies, transparent reporting) before the public meeting, not after it. Then fiber, because AI workloads are networked as well as power-hungry, and energy-rich sites with diverse fiber carry a different value profile than generic dirt. Then workforce, where Ohio's industrial memory — maintenance, energy, logistics, manufacturing, trades, utilities, agriculture — is worth more on a powered campus than in a pure software economy.

And then the principle developers underrate until it kills them: community posture. A site can have gas, grid, fiber, and water and still fail if the town believes the project takes more than it returns. EnergiAcres' Mansfield thesis is built on exactly this point. Mansfield is a town near where I grew up — five closed steel mills, outmigration, a thinned tax base — and it is our flagship proof that energy-first siting can be turned into something a data-center operator and a greenhouse operator would both choose, and that a community would welcome rather than resist. The right frame for the state is not "bring us a data center." It is "bring us a new infrastructure platform."

Hosting compute is the commodity layer. The higher-value prize is to become the operating template for community-aligned AI infrastructure.

That template is a repeatable order of operations: screen first for energy, then land, then water, then fiber, then entitlement, then community posture, then tenant fit. In most development categories that sequence would look strange. In AI infrastructure it is the only order that makes sense. Ohio should not try to be Northern Virginia, or Texas, or the cheapest dirt in the auction. Its advantage is its own: gas, legacy industrial corridors, manufacturing DNA, underused land near real infrastructure, and towns that already understand what happens when industry leaves — communities that do not need to be sold a fantasy, only shown a better bargain. The next industrial wave will be won by places that can say yes without being naïve. Ohio can be one of them, but only if it builds from first principles.