Sen. Mark Warner has a blunt answer for communities bracing for AI-driven job losses: make the booming data center industry pay a “pound of flesh” to fund the transition. In private and public remarks in Washington this week, the Virginia Democrat sketched a plan to tax data centers powering generative AI and route the proceeds into worker support, local benefits, and skills programs—arguing it is the fastest way to match the scale and speed of disruption already rattling entry-level markets.
Warner’s pitch lands amid mounting anxiety. U.S. entry-level job postings have dropped about 35% since 2023, Big Tech layoffs have rippled through multiple sectors, and AI leaders themselves now caution about near-term labor displacement. Warner cited fresh anecdotes: a venture investor writing down traditional software bets after seeing rapid progress from Anthropic’s Claude, and a major law firm slowing first-year associate hiring as AI takes on routine legal tasks.
A Tax for the Transition to Support Workers and Communities
The senator’s core idea is straightforward: harness a portion of data center gains to underwrite tangible local benefits where the facilities are built. That could mean tuition-free upskilling for displaced workers, scholarships into in-demand roles like nursing, and community impact funds that directly improve housing, transit, or schools.
Warner points to a pragmatic precedent in his backyard. Henrico County, Virginia, leveraged tax revenue tied to data centers to jumpstart affordable housing—proof, he argues, that concentrated digital infrastructure can yield immediate, visible returns for neighbors rather than just far-off corporate balance sheets.
The political logic is as important as the economics. By connecting AI’s infrastructure to on-the-ground benefits, Warner aims to blunt resentment toward facilities that can be noisy, resource-intensive, and power-hungry. The goal: keep the buildout on track while ensuring communities don’t feel like collateral damage of an AI race they didn’t choose.
Public Backlash Creates Urgency for Local Benefits
Public sentiment is souring. An NBC News poll recently found 46% of registered voters view AI negatively versus 26% positively—a rare tech backlash with electoral implications. In Virginia, where Northern Virginia is the world’s largest data center market by capacity according to CBRE, lawmakers are even weighing the repeal of generous tax breaks that cost state and local budgets nearly $2 billion annually.
Nationally, opposition is hardening. This week, Sen. Bernie Sanders and Rep. Alexandria Ocasio-Cortez introduced a bill calling for a moratorium on data centers. Warner rejects that approach, warning that halting U.S. construction would simply shift advantage abroad. He argues it is better to channel a slice of industry profits into community priorities than to freeze the infrastructure that underpins AI competitiveness.
The broader mood is unmistakable: without clear, local wins, communities will reach for pitchforks. A visible, earmarked revenue stream could lower the temperature and replace blanket opposition with conditional support.
What Would Be Taxed and How Data Centers Are Assessed
Design matters. Policymakers have a menu of options:
- Modest surcharges on electricity consumption above a threshold
- Impact fees tied to water use
- Incremental property tax assessments on high-density server space
- Targeted excises on compute used for large-scale AI training
Each tool carries trade-offs for grid reliability, location decisions, and administrative complexity.
One principle is gaining traction among analysts: tie levies to measurable externalities and recycle proceeds locally. That could fund grid upgrades, noise abatement, water recycling, and workforce programs within the host county—mirroring how severance taxes support communities near extractive industries. Clear earmarks also help maintain public trust that funds won’t disappear into general revenue.
Crucially, any framework must avoid whiplash. If rates swing wildly or differ too sharply across counties, hyperscalers will leapfrog jurisdictions. A federal floor with state flexibility—or regional compacts in high-growth corridors—could reduce the risk of a race to the bottom while preserving local control.
High Stakes For The Economy And The Grid
Behind the tax debate lies a larger infrastructure crunch. The International Energy Agency projects global data center electricity demand could roughly double by 2026, with AI a prime driver. U.S. energy officials expect data centers to strain specific grids in fast-growing hubs, forcing new generation, transmission, and efficiency investments on an accelerated timetable.
Northern Virginia remains the epicenter, but the footprint is spreading to the Midwest, the Southeast, and power-abundant regions near hydro and nuclear assets. As facilities proliferate, communities want a clear bargain: jobs and local amenities in exchange for land use, noise, and higher peak loads. Warner’s framework tries to formalize that bargain without slamming the brakes.
The labor calculus is equally unsettled. Goldman Sachs Research has estimated that AI could expose hundreds of millions of roles globally to automation, even as new categories emerge. Other analyses suggest adoption lags and task reconfiguration may temper near-term displacement. Warner’s message splits the difference: act now on retraining and community benefits, rather than waiting for perfect attribution of which job was “lost” to which model.
Open Questions Before a Bill Arrives in Congress
Warner has not yet introduced legislation, and key questions remain:
- What revenue baseline will meaningfully fund reskilling at scale?
- How will tax design recognize facilities that run on low-carbon power or invest in local infrastructure?
- Can regulators distinguish energy-thirsty AI training from routine enterprise compute without burdensome reporting?
- How will states coordinate to prevent tax arbitrage without stifling investment?
Still, the contours are clear. Data centers are the physical engine of the AI economy. If that engine is going to rev faster, Warner wants it to tow some of the social load—financing the worker pathways and community upgrades that make the transition politically and economically sustainable. In a season of AI skepticism, that may be the price of keeping the buildout on track.