Data Centers Got Crushed On The Ballot In The 2026 Midterms

Voters across the United States delivered a clear message on November 3, 2026: most of them do not want a large data center built next door without strong limits. Roughly 40 to 41 local measures on data centers were on ballots in about 10 to 13 states, and at least 33 of them passed in favor of restrictions. Some early counts put the number even higher, at 37 or 38 of the first measures called. The data centers crushed ballot 2026 midterms story was not limited to one region or one party. It ran from Florida to California, from Maine to Ohio, and it included outright bans, temporary pauses, new rules and caps on size.

Quick Facts

Data Centers Crushed Ballot 2026 Midterms: What The Numbers Show

The headline number is striking, but the detail matters too. These were not national votes. They were local votes, in towns and small cities, where residents were asked a direct question about building in their own backyard. In many places the question was whether to ban a facility, pause new projects while rules were written, or set a firm limit on how large a facility could be. Voters answered yes to limits again and again.

Restrictions passed in conservative areas and liberal areas, in suburbs and small towns. The common thread was a local concern about power, water, noise, traffic and what a huge industrial building would mean for a neighborhood. The one clear exception was Harwood, North Dakota, a town of under 1,000 people, where about 430 voters took part and about 75 percent voted against a restriction. Everywhere else that had been called, limits won.

Ohio Led The Way, And Every Measure There Passed

Ohio had the most data center measures on the ballot, with at least 18 to 21 communities voting. According to USA Today and Route Fifty, all of them were approved. That alone would make Ohio the center of this story, but the individual results show how voters used different tools for the same goal.

In Defiance, Ohio, voters banned data centers over 25 megawatts by a vote of 63 to 37. That cap draws a line between a small local facility and the huge campuses built to power artificial intelligence, and it targets the very large ones that raise the biggest worries about electricity and water.

The statewide result sent a signal beyond Ohio: a data center proposal there cannot be treated as a routine approval. Mayors in Ohio and Oklahoma were even recalled over data center backlash, showing anger about decisions leaders had already made, not only about future projects.

From Florida To California, Bans And Pauses Won Big

The results were just as strong on the coasts. In Edgewater, Florida, voters approved a charter ban on data centers with more than 75 to 77 percent of the vote. A charter ban is a strong step because it writes the rule into the basic governing document of the city, not just into a single ordinance that a later council could quietly change.

In Alhambra, California, voters banned data center facilities with about 82 percent of the vote. That is a landslide by any normal election standard. Monterey Park, California, had already taken a similar step with an earlier ban, and its experience was often mentioned in coverage of the new California votes. On the other side of the country, Bloomington, Illinois, approved a moratorium, which is a temporary pause on new projects while a city studies the issue and writes rules.

Peru and Wiscasset, Maine, also approved moratoriums. A moratorium does not settle the issue forever. It buys time. For voters who felt that projects were moving faster than local rules could handle, that pause was the point. They wanted a chance to study power demand, water use and local costs before a single large building was approved.

At least 10 municipalities banned large centers or centers over 25 megawatts, and six to seven banned data centers outright.

Why Voters Said No: Power Bills, Water And Jobs

Ask voters why they opposed a data center and three worries come up again and again: power bills, water use and jobs. Many residents fear that a huge facility will strain the local power grid and push up electricity costs for homes and small businesses. Even when a developer promises to pay for its own power, neighbors worry about who pays for new lines, new substations and backup capacity.

Water is the second worry. Large data centers use water to cool their servers, and in dry states residents think first of homes, farms and local wells. Supporters say modern cooling uses far less water, while critics want clear public numbers before approval, not promises after the building is finished.

Jobs are the third issue. Supporters point to construction jobs, tax revenue and investment. Opponents answer that a finished data center employs far fewer people than a factory of the same size, and that most permanent jobs are highly skilled roles that may not go to local residents. For many voters, that trade did not add up.

Voters also asked why their town should carry the costs of artificial intelligence that benefits users around the world.

What The Exit Polls And Candidate Results Add

The local ballot results did not happen in isolation. NBC News exit polls found that 56 percent of voters said AI and data centers were a crucial or major factor in their vote for Congress. About 76 percent said they would oppose a new data center in their own community. Those numbers help explain why the local measures passed so widely. This was not a niche issue that only activists cared about. It was on voters minds when they chose national leaders too.

Candidates who favored guardrails also won in several states. In Texas, governor elect Gina Hinojosa has called for a statewide moratorium, and candidates favoring guardrails won in Michigan, Arizona, South Carolina and Ohio. Voters weigh many issues at once, but the pattern fits the ballot results.

The data centers crushed ballot 2026 midterms result therefore has two layers. The first layer is local and direct: bans, pauses and caps in town after town. The second layer is political: a signal to state leaders that voters want rules before the next wave of building, not after it. State lawmakers in places with many proposals will now face pressure to set statewide standards for power, water disclosure and local approval.

What Happens Next: Advisory Votes, New Rules And More Ballots

Not every measure from November 3 has the same legal force. Some measures are advisory, which means they express the will of voters but do not by themselves change the law. Others are binding bans or ordinances that local officials must follow. Voters and developers alike will need to read the fine print in each town to know what actually changes on the ground.

More votes are still to come. Arkansas and North Dakota are expected to hold further votes next month, and other towns may place measures on future ballots after seeing this year’s results. Expect future proposals to carry more detail on power, water, noise and community payments, because a vague promise is now less likely to pass.

Demand for computing power is not going away, so the likely result is not an end to data centers. It is a shift toward smaller facilities, stricter rules, better disclosure and sites chosen with more local consent.

More from The Akron Magazine: for the wider election picture, read our report on 2026 midterm results, our coverage of global stocks sliding after the Democratic sweep as AI data center fears hit tech shares, and our story on the Dow jumping more than 1,100 points as Wall Street cheered divided government.

Frequently Asked Questions

How many data center measures were on the ballot in 2026?

Roughly 40 to 41 local measures on data centers were on ballots on November 3, 2026, across about 10 to 13 states. At least 33 passed in favor of restrictions, and some early counts put the figure at 37 to 38 of the first measures called. Counts differ slightly because some results were still uncalled when trackers reported.

Did any town vote against restricting data centers?

Yes, one notable exception was Harwood, North Dakota. About 430 voters took part, and about 75 percent voted against a proposed restriction. Harwood has under 1,000 residents, and it was the only place in early counts where voters rejected a limit.

What is the difference between a ban and a moratorium?

A ban forbids a type of facility, either outright or above a size such as 25 megawatts. A moratorium is a temporary pause on new projects while a city studies the issue and writes permanent rules. Both were used widely in 2026, and both were approved by voters in the places where they appeared.

Will these votes stop data centers from being built?

No, not everywhere and not entirely. Some measures are advisory rather than binding, and demand for computing power remains high. The more likely effect is stricter local rules, smaller projects, clearer disclosure on power and water, and more towns requiring real local consent before a large center is approved.

Sources

USA Today: Data centers and the midterm elections

Route Fifty: Voters reject data centers in local ballot initiatives

Gizmodo: Data centers on the ballot

Final Thoughts

The data centers crushed ballot 2026 midterms result will be remembered as the night local voters took control of a national tech debate. They did not vote on artificial intelligence in the abstract. They voted on a building, a power line and a water bill in their own town, and in town after town they chose limits. That choice was not anti technology. It was a demand for rules, honesty about costs and a real say before construction starts.

Developers and state leaders now have a clear task: show their numbers, respect local limits and earn consent rather than assume it. The votes still to come in Arkansas and North Dakota will test whether that lesson has been learned.