Overview of The Biggest Loser of the Midterms? Data Centers
This episode of The Journal examines how data centers became an election issue in the U.S. midterms—and why the political backlash could meaningfully slow the AI boom. Voters across several states and localities rejected or restricted new data center development, signaling rising public resistance to the land, water, power, and quality-of-life costs of AI infrastructure. The episode explores how this may constrain AI companies’ growth, pressure markets, and force a rethink of where and how the industry builds compute capacity.
Why Data Centers Became a Political Target
Data centers have become a visible symbol of AI expansion, and many communities now see them as a burden rather than a benefit.
Main complaints from voters and local residents
- Higher electricity costs
- Heavy water usage
- Noise and environmental disruption
- Few permanent jobs relative to the scale of investment
- Loss of farmland and changes to the local landscape
- Lack of transparency from companies, including NDAs in some development talks
Political impact
- Democrats made data centers a campaign issue in several places
- Voters in states like Texas, Kansas, Georgia, and Michigan backed anti-data-center candidates or measures
- More than a dozen towns in Ohio voted to halt or restrict new developments
What the Midterm Results Mean for AI Companies
The episode argues that the election results were a major wake-up call for AI firms that had assumed most of their projects would move forward.
The core concern: compute shortages
AI companies depend on continual growth in computing capacity to:
- Serve more users
- Train new models
- Stay competitive in a fast-moving market
If data center construction slows, companies may hit a ceiling on how much they can grow.
Why this matters financially
- Revenue growth is tightly linked to compute expansion
- Slower capacity growth could hurt companies like OpenAI and Anthropic
- A slowdown could also affect IPO prospects and broader investor confidence
Market Reaction
The market interpreted the election results as a real threat to the AI buildout.
Immediate fallout
- Shares tied to AI infrastructure fell sharply
- Chipmakers, data center suppliers, and “picks-and-shovels” companies were hit
- The Nasdaq was dragged down as much as 5% before rebounding somewhat
Broader worry
A slowdown in AI infrastructure could ripple outward into:
- Semiconductor companies
- Supply-chain businesses
- Cloud and “neocloud” providers
- Companies and sectors indirectly riding the AI boom
How AI Companies May Respond
With local resistance rising, companies are looking for ways to build without triggering the same backlash.
Possible workarounds
- Retrofitting existing buildings into data centers
- Building smaller, less disruptive facilities
- Seeking sites and strategies that avoid new permits or major construction
- Shifting some investment outside the U.S.
Strategic shift
Meta’s Dina Powell McCormick suggested that every hyperscaler is now considering overseas expansion as a hedge against U.S. political and regulatory delays.
Broader Takeaways
The episode frames data centers as the physical bottleneck of the AI economy—and now, its political liability.
Key insight
Because data centers require permits, utilities, land, and often tax incentives, they are unusually exposed to local politics. That makes them one of the first points where public frustration with tech has translated into real electoral consequences.
Potential upside for the industry
Some in tech may quietly welcome a slowdown, since it could:
- Force more disciplined spending
- Cool the current arms race for capacity
- Reduce overbuilding before revenue catches up
Bottom Line
The midterms showed that voters are willing to push back on the infrastructure behind AI, not just the technology itself. If that resistance continues, it could slow U.S. data center expansion, reshape where AI companies build, and create a serious constraint on the industry’s growth trajectory.