By: Mitch Heimpel
I have unfortunate news for all of us. The Data Centre Wars are here.
In the United States midterms earlier this month, ballot initiatives to reject data centres got the backing of millions of voters. Incumbents pledging to oppose data centres, like Democratic rising star Jon Ossoff, took direct aim at the data centre build out, saying that tech CEOs were trying to “dig bunkers for themselves.”
Following the midterms, I did an analysis of the 50 most controversial data centre projects in the U.S. (This is how I have fun. Don’t judge me.) My standards for controversy were pretty modest, and entirely based on my prior political experience with wind farms — more on that later. I used the size of a Facebook group opposing the project, size of any online petitions opposing the projects, number of media stories about the project, and any active Reddit or sub-Reddit forums about the project as a proxy for how controversial said project actually was in real life.
I did that because the challenge with assessing the relative effectiveness of these kinds of local political efforts — or moral panics, depending on one’s point of view — is that they’re loud but not always particularly organized or effective. I wanted to know how many people were talking about these, where they were gathering online, and, ultimately, whether they were effective.
What did the midterms show us? In the counties represented by those 50 most-controversial projects — 46 counties in total across the United States, since some had multiple projects — the swing to the Democrats was 17 points. Across the country, it was only 14.7 points.
Now, that’s not a huge difference, and we can’t attribute all of it to data centres. Still, there are examples worth noting. One of the biggest projects in the United States is the xAI Colossus Project in Southaven, Mississippi. DeSoto County, where Southaven is located, went for Trump by 24 points in 2024. When the race is finally certified, it looks like it will go for Democratic Senate candidate, Scott Colom, by six points. That’s a 30-point swing in the closest Senate race in the country, in just the county that is expected to house one of America’s biggest data centre projects.
This is not a uniquely American development. Something similar could be seen in Ontario during the recent province-wide municipal elections. Oakville and Mississauga, both municipalities facing high-profile data centre development, turfed incumbent mayors. In Oakville’s case, it turfed an almost two-decade incumbent. In smaller municipalities like Centre Wellington or Quinte West, data centres became a live election issue. Both those municipalities will have new mayors this term.
This is something I have now seen first-hand. I worked on a campaign during the recent elections, namely, the above-mentioned Quinte West race. Data centres were an issue that we heard about constantly. I don’t know if a day went by when it didn’t come up while we were campaigning. My own personal sense, for what it’s worth, is that while it was hard to put a number on how many people were actively opposed to the project, it was enough to impress local officials, who were surprised by the scale of the of public response the proposal generated. However many of them there were, they really cared about the issue.
This is by no means limited to Ontario. Data centres operate coast-to-coast already. Many went unnoticed and opened without issue. New ones aren’t going to get in under the radar. They’re going to become lighting rods. Communications are enabling a level of organization among opponents that simply hasn’t existed before.
I confess that I was surprised to see this, but it did remind me of something I saw earlier in my career. Many moons ago, I was a political staffer at Queen’s Park, during the era of the Green Energy Act and its emphasis on solar and wind farms. A lot of what I’m starting to see bubbling about data centres reminds me of those wild times.
Opposition to data centres (like wind farms before them) can be understood as breaking out roughly into three groups. There is a small group who are genuinely aggrieved by legitimate process concerns. It’s not unusual for planned projects to try and ram themselves through quickly, or bypass (often dysfunctional) local zoning committees and policies, and that can turn people off. And sometimes there truly are legitimate local concerns about the specific project, or the specific location chosen for a project. Not every objection is pure, knee-jerk NIMBYism.
The next group of objections are quasi-legitimate. They might be policy-based. Usually, it’s a response to what the locals perceive as a “lack of consultation.” If that seems like a nebulous term, it is so on purpose. Some people are legitimately concerned about it. Others are just cynically using the approvals process to try and smother projects they dislike to death under reams of red tape. Both sides — the well-intentioned and the cynical — will grab onto published info about data centres elsewhere, even if the projects are nothing like what’s being proposed in their community, and base their arguments around that. This is exacerbated in Canada because so little information about a project is proactively disclosed to the public. There’s rarely good info with which to combat the bad. So many of these objections, whether good faith or not, often veer into misinformation.
The final level of concern is essentially something akin to a moral panic. A generation ago, with wind turbines, people were terrified about “wind turbine syndrome” — purportedly a combination of low-frequency noise and vibrations that was alleged to cause a myriad of symptoms or illnesses. Dead birds were often frequently cited, as well. In the case of data centres, it’s fear over local water resources being used up, or contaminated, by data centres — specifically, the large cooling systems required to operate them.
When people think you’re contaminating the water, or causing an outbreak of illnesses or annihilating whole flocks of birds, they’re no longer arguing about process. They’re arguing the morality of building the thing at all. Project proponents often recoil from a debate like that.
And that recoiling can be a sight to behold. In Ontario, back when I was a political staffer, when the backlash against wind farms came, it came fast. Local planning powers had been overridden because the then-Liberal provincial governments were all-in on fighting climate change. That lasted until it didn’t — the public support just collapsed. We ended up cancelling and in some cases actually demolishing, at considerable expense, half-completed projects, because the political costs of finishing them had simply become untenable. In the end, we spent real money to not build windmills.
It would be better to not do that again. Data centres are a complicated issue, and there’s a range of perfectly legitimate views. Personally, I think we need them. Canadians are voracious users of AI, consuming much more global compute usage than our population alone would suggest. We clearly enjoy using the stuff, but we don’t want the centres in our communities. This puts us in an awkward spot. We can rely on foreign-based data centres, which poses all kinds of sovereignty problems (especially considering the government’s stated desire for Canada to be more independent), or, we can figure where to put domestic data centres. On balance, the latter is probably the right call.
What is that going to mean for our politics? Every data centre project is now a communications project with a CPU attached. Tech companies, and other proponents of data centres, need to realize this, and communicate responsibly and effectively. In yet another reminder of the wind-farm wars, proponents are overly relying on what I call “inevitability arguments” to make their case. “We’re going to need to put them somewhere,” “We all use AI,” “The technological revolution is too important to miss out on” — you hear variations of that everywhere. To concerned locals, though, this sounds a lot like “You don’t understand the problem we’re urgently trying to solve, so get out of our way.”
With wind farms, the problem was climate change; with data centres, it’s the AI revolution and all the threats and opportunities it entails. If tech companies try to combat online organization with narratives about the inevitable march of technological progress, they are going to learn very quickly that Facebook and Reddit offer ready-made organizing platforms with a nasty tendency to transform online grievances into ballot-box outcomes. Politicians can be easily swayed by a well-run online campaign, and if that happens, your project is in real trouble. It’s one thing to have an online community opposing your project. It’s another to have a mayor, a council, an MP, an MLA , or a cabinet minister, elected by that online community, standing in your way.
Canada probably really does need to develop our own domestic compute capacity here. That is going to be mean data centres on our soil, under our law. It would be nice to think that we can do this. But as a scarred veteran of those old windmill campaigns, I have to admit I’m nervous. The data centre wars are here, in our politics, right now, and I’m not sure we’re ready for them.
Mitch Heimpel is the vice president of government relations at Texture Communications and has worked for Conservative leaders and cabinet ministers federally and provincially.
The Line is entirely reader and advertiser funded. No federal subsidies, no bailouts. If you value our work, please consider supporting us by subscribing or making a donation. Donations are not subscriptions and do not unlock paywalled content, but they help keep The Line independent
To contact The Line with a general inquiry or comment, please email info@readtheline.ca. For other ways to connect with us or to follow us on social media, please see our LinkTree.





I covered a wind project in SW Ontario for a small local newspaper. It was shocking to me to see very “nature-oriented” people (hippies) oppose a renewable energy project to the point where they made a staffer cry in a public meeting. No one notices or cares about the turbines now.
Canader eh!! oppose anything that will make money, data centers, pipelines. What next??? Why people won't invest here!!