Ohio Data Centers Need Standards, Not a Place in the Constitution

There is now a proposed constitutional amendment for an Ohio data center ban to prohibit the construction of data centers. That should stop people cold. A state constitution is where you define the structure of government and protect fundamental rights. It is not where you hard-code a ban on a category of digital infrastructure because the public debate around it has become overheated.

Ohio data centers deserve scrutiny. They use real power. Some use real water. They impose real demands on local infrastructure. But that is an argument for competent planning, not constitutional prohibition. Ohio should regulate them seriously, site them carefully, and demand clear standards. What it should not do is take itself out of the game just as cloud and AI infrastructure become more central to economic growth.

“Ohio data centers need planning and standards. They do not belong in the state constitution as a prohibited industry.”

Start with the obvious: data centers are already part of Ohio’s economy

This debate is often framed as if Ohio is deciding whether to allow some entirely new type of development into the state. That is false. Ohio is already in this business. AWS operates its US East (Ohio) region with three Availability Zones, and central Ohio already hosts Amazon data center infrastructure. This is not speculation about some distant future. It is existing infrastructure, on the ground, in the state.

That matters because the question is not whether Ohio will ever encounter data centers. It already has. The real question is whether Ohio intends to manage growth intelligently or respond to it with a constitutional overreaction.

Yes, data centers use a lot of electricity. That is exactly why planning matters

Critics are right about one thing. Data centers are major power users. The U.S. Department of Energy reported that American data centers consumed about 176 terawatt-hours in 2023, roughly 4.4% of total U.S. electricity demand. DOE also projects substantial growth from there, potentially reaching 325 to 580 terawatt-hours by 2028.

Those are not trivial numbers. They mean states need to think seriously about generation, transmission, local distribution, reliability, and cost allocation. They also mean communities should ask hard questions before approving projects. How large is the load? What upgrades are required? Who pays? What happens to local reliability? Those are real governance questions.

But this is where the ban argument falls apart. Large electricity demand is a reason to improve utility planning. It is not a reason to amend the constitution.

“A large electric load is a planning problem. It is not a constitutional crisis.”

Water use is real too, but the public discussion is often sloppy

Water is the other topic that gets discussed in broad, often misleading claims. Some data centers do use substantial water, particularly those using evaporative cooling. That is a legitimate concern. But water use is not a fixed number that applies equally to every facility in every location.

A January 2026 fact sheet from the Environmental Law Institute, summarizing recent research, estimates that U.S. data centers directly consumed about 66 billion liters of water in 2023. It also notes that water demand varies significantly by cooling design, climate, and operating conditions. Some newer cooling approaches, including direct liquid cooling and other closed-loop methods, can significantly reduce on-site water consumption.

That is the point policymakers should focus on. What cooling system is being proposed? What is the expected annual water profile? What are the local water constraints? What standards or disclosures should apply? Those are serious questions. “Data centers use water” is true, but it is not a complete argument.

The economic case is real, but it should be stated honestly

Data centers are frequently oversold as giant permanent job machines. That is not the best case for them, and people should stop pretending otherwise. A single facility may employ fewer full-time on-site workers than the public expects. Critics are right to question exaggerated claims.

But the economic case does not disappear just because the simplistic jobs pitch is weak. The Ohio Chamber Foundation’s 2025 study estimates that data centers supported more than $1 billion in annual state and local tax revenue in 2024 and were associated with broad economic effects across construction, operations, and supply chains. The same study says the sector has attracted more than $40 billion in private capital investment in the state.

AWS’s own Ohio materials state that the company contributed an estimated $2.2 billion in GDP from 2015 through 2022, supported an estimated 3,550 average annual full-time-equivalent jobs at local vendors, and invested $6.3 billion in Ohio over that period. Company-provided numbers should always be read with the usual caution, but they still underscore the basic point: this is not imaginary economic activity.

“The question is not whether every data center is good. The question is whether Ohio is serious enough to govern them well.”

Data centers also matter because they are enabling infrastructure

This is the part critics often miss. Data centers are not just buildings full of servers. They are enabling infrastructure for cloud services, enterprise software, logistics, finance, healthcare systems, media delivery, and increasingly AI workloads. States that can support this infrastructure credibly are more likely to attract adjacent investment in the industries that sit on top of it.

That spillover is not automatic. It depends on power, fiber, workforce, local policy, and whether state leaders know how to convert infrastructure presence into broader growth. Brookings has made this point directly: the payoff is strongest when communities turn data center investment into a larger ecosystem strategy rather than treating each facility as an isolated project.

That is why a constitutional ban is such a self-defeating move. It does not merely block a category of construction. It tells the broader technology economy that Ohio is willing to remove itself from critical infrastructure markets by referendum.

A serious state distinguishes between regulation and prohibition

This should not be a hard concept. Ohio can be demanding without being self-destructive.

It can require transparent disclosure of expected electricity demand and water use. It can insist on credible utility planning before approvals are granted. It can scrutinize incentives. It can impose siting standards, noise controls, environmental conditions, and clear reporting requirements. It can decide that some projects are bad fits for some communities. All of that is normal governance.

What it should not do is confuse frustration with policy. A constitution is the wrong tool for calibrating industrial siting decisions. Once you start using it that way, you are no longer governing. You are freezing policy in place because you do not trust your own institutions to do their jobs.

Ohio should stay competitive without pretending impacts do not exist

There is no need for boosterism here. Data centers are not impact-free. They are large industrial users of power. Some are meaningful users of water. They require competent oversight and, in some cases, skepticism.

But none of that supports a constitutional ban.

The sensible position is straightforward. Ohio should keep standards high, require serious planning, and evaluate projects on facts rather than panic. It should stay open to a form of infrastructure that already exists in the state and will matter even more over the next decade. If a project is badly sited or badly designed, reject it. If it is responsibly planned and economically sound, permit it.

This is what a serious state does. It governs. It does not write itself out of the future.

---

The Ohio Prohibition of Data Center Construction Amendment may appear on the ballot in Ohio as an initiated constitutional amendment on November 3, 2026.