Lawmakers Push Back on Big Tech’s Data Center Expansion
The rapid build-out of AI-driven data centers across the United States is meeting growing resistance from federal and state lawmakers concerned about rising electricity bills and grid strain.
This week, Senators Josh Hawley and Richard Blumenthal introduced the first bipartisan federal bill aimed at preventing data center power consumption from being passed through to everyday ratepayers.
At the state level, New York became at least the sixth state to propose legislation that would temporarily pause new data center construction. Similar bills have emerged in Georgia, Virginia, Oklahoma, Vermont, and Maryland.
The concern is simple: AI infrastructure growth is outpacing the regulatory framework designed to manage it.
Hyperscalers Are Spending at Historic Levels
The four major hyperscalers —
- Microsoft
- Alphabet
- Amazon
- Meta
— are projected to spend roughly $650 billion this year on AI-related infrastructure.
A substantial portion of that capital is flowing into:
- Massive server warehouses
- Grid upgrades
- Power interconnection infrastructure
- Cooling systems
According to estimates from Lawrence Berkeley National Laboratory, U.S. data center electricity demand doubled between 2018 and 2024 and could triple again by 2028.
In PJM Interconnection territory — the nation’s largest grid operator — capacity prices jumped from $28.92 per megawatt-day in 2024–2025 to $329.17 per megawatt-day for 2026–2027.
That surge feeds directly into utility cost structures.
Electricity and Water Concerns Intensify
Beyond power consumption, megasize hyperscaler facilities require enormous water volumes for cooling.
Forecasts suggest hyperscale data centers could use more than 150 billion gallons of water over a multi-year span — comparable to the annual usage of millions of households.
Lawmakers argue that:
- Ratepayers should not subsidize AI infrastructure
- Grid upgrades should be developer-funded
- Local communities deserve environmental protections
New proposals increasingly require developers to fully fund power interconnections, grid upgrades, and environmental studies.
Tech Companies Respond
AI developers have begun pledging cost-offset measures.
Microsoft has stated it will pay utility rates sufficient to fully cover energy costs and replenish more water than its facilities consume.
Amazon says it has reduced water usage per computing unit by roughly 40% since 2021.
Anthropic recently announced it would fund 100% of required grid upgrades for its facilities.
OpenAI has made similar commitments.
Still, regulatory pressure continues to build.
Economic Trade-Offs: Jobs vs. Tax Breaks
The debate is not one-sided.
In Georgia alone, data center development created:
- 8,500+ construction jobs
- 1,600+ permanent operations jobs
- Over $1 billion in economic impact
However, tax exemptions granted to attract developers have reduced state revenue by hundreds of millions of dollars.
Virginia, home to the largest concentration of data centers in the U.S., faces mounting scrutiny over billions in lost tax revenue due to incentives.
For states, the calculus is complex:
- Short-term economic growth
- Long-term grid strain
- Ratepayer exposure
- Environmental impact
Grid Operators Face Planning Risk
Another issue complicating the situation: not all proposed data centers will actually be built.
AI chip shortages, construction labor constraints, and power access bottlenecks mean the development pipeline far exceeds what can realistically be delivered.
When utilities begin building transmission infrastructure for projects that later stall, costs often fall back on ratepayers.
Many new legislative proposals now focus heavily on risk allocation, requiring developers to post deposits, fund studies, and guarantee long-term power commitments.
Site Selection Is Already Shifting
As opposition grows in traditional hyperscaler hubs like Northern Virginia, companies are increasingly targeting second-tier states with available generation capacity, including Kentucky and Indiana.
Meta is currently developing a 1-gigawatt data center in Indiana — a scale equivalent to powering hundreds of thousands of homes.
Developers appear to be quietly repositioning toward jurisdictions offering:
- Faster permitting
- Lower political resistance
- Greater grid flexibility
WSA Take
The AI boom isn’t just a software story — it’s an infrastructure shock.
Electric grids were not designed for single-sector load surges of this magnitude. As hyperscalers race to dominate artificial intelligence, policymakers are scrambling to prevent consumers from absorbing the cost.
Expect three major trends:
- More cost-allocation reforms
- Increased grid upgrade requirements
- Geographic shifts toward more permissive states
The AI arms race continues — but it is now colliding with the physical limits of the power grid.
And that collision is becoming political.
Disclaimer
WallStAccess is a financial media platform providing market commentary and analysis for informational and educational purposes only. This content does not constitute investment advice, a recommendation, or an offer to buy or sell any securities. Readers should conduct their own research or consult a licensed financial professional before making investment decisions.
