AI Job Losses Are Real — And One Senator Wants Data Centers to Pay
Artificial intelligence is quietly reshaping the job market, and the warning signs are no longer subtle. Entry-level job postings in the United States have dropped 35% since 2023. Major law firms are freezing junior associate hiring. Venture capitalists are writing software investments down to zero. And now, a sitting U.S. senator is pushing a bold idea to make the companies powering this shift pay for it — starting with the data centers behind the AI boom.
 |
| Credit: Getty Images |
The Fear of AI Job Displacement Is No Longer Hypothetical
For years, talk of AI replacing human workers felt distant — a futuristic concern dressed up in think pieces and TED talks. Not anymore. The economic reality of AI-driven job displacement is arriving faster than most policymakers anticipated, and it is hitting hardest at the bottom of the career ladder.
Entry-level roles, once considered safe entry points into professional careers, are disappearing. A major law firm recently confirmed it is no longer hiring first-year associates because AI can now handle much of the foundational legal work those young professionals used to do. A prominent venture capitalist told lawmakers he is writing down software investment valuations to near zero, citing rapid advances in AI coding capabilities. These are not forecasts. They are decisions already being made in boardrooms across America.
Senator Mark Warner of Virginia summed it up bluntly: the fear of AI job loss is "palpable." Even as some AI companies argue the data does not yet show widespread displacement, the anxiety is already reshaping political conversations at the highest levels of government.
Who Should Pay for the AI Transition — And Why Data Centers Are in the Crosshairs
When a new technology creates winners and losers, the question of who bears the cost of the transition is rarely answered fairly. Senator Warner is trying to get ahead of that question before the economic pain deepens.
His proposal: tax data centers — the physical infrastructure that makes AI possible — and direct that revenue toward helping workers navigate the shift. The idea has not yet been introduced as formal legislation, but Warner has been testing the concept publicly, including at a recent AI summit in Washington, D.C. He raised a pointed question that cuts to the heart of the debate: should chip manufacturers bear the cost? Should the companies building large language models? Should the banks and financial giants using AI to eliminate analyst positions?
Warner's answer is pragmatic: the easiest and most logical place to extract accountability is from the data centers. They are the visible, physical, taxable presence that communities are being asked to welcome — often at significant cost to local infrastructure, electricity grids, water supplies, and quality of life.
Data Centers Are Already Facing a Community Backlash
The political heat around data centers is not just about jobs. Across the country, residents are pushing back against the facilities for noise, pollution, and the strain they place on local power grids and water systems. That anger has been building for months, and it is now breaking into legislative action.
Prominent progressive lawmakers recently introduced a bill calling for a full moratorium on new data center construction. The proposal reflects genuine frustration at the grassroots level — a resentment rooted not just in environmental concerns, but in a deeper anxiety. Why should communities absorb the disruption of having a massive facility in their backyard if the technology it powers is the same technology threatening their livelihoods?
Senator Warner does not support the moratorium. He argued that halting data center construction would simply hand a competitive advantage to China, a country racing to dominate global AI development. But he acknowledged that communities deserve something meaningful in return for what they are being asked to absorb. His phrase for it is direct: a "pound of flesh."
What a Data Center Tax Could Actually Look Like
The concept Warner is floating is not without real-world precedent. Henrico County in Virginia used tax revenue generated by a local data center to launch a new affordable housing initiative — a tangible, community-level benefit that turned infrastructure controversy into local investment.
Warner envisions a similar model scaled up. Data center tax revenue could fund nursing training programs, AI upskilling initiatives, and workforce transition support for workers whose roles have been eliminated or diminished by automation. The goal is to create a direct, visible link between the facilities communities are hosting and the economic benefits those communities receive in return.
This is not charity. It is a structural correction — a recognition that the companies profiting most from AI infrastructure should contribute to managing the disruption their technology creates. Warner framed it as an obligation the industry has long avoided and can no longer ignore.
Public Opinion Is Turning Against AI — And Fast
The political urgency behind Warner's proposal gains sharper context when you look at how Americans are actually feeling about artificial intelligence right now. A recent national poll found that AI holds a lower public approval rating than Immigration and Customs Enforcement — a striking data point given how polarizing immigration enforcement has become. Forty-six percent of registered voters view AI negatively. Only 26% view it positively.
That sentiment is already producing real policy consequences. In Virginia — home to one of the largest data center markets in the world — lawmakers are considering repealing state tax breaks for data center construction. Those breaks currently cost the state and local governments nearly two billion dollars per year in lost revenue. Other states are watching closely and may follow.
AI and data centers, as Warner noted, are easy to demonize. That is partly because the benefits of AI are abstract and diffuse while the costs — job losses, rising electricity bills, environmental strain — are immediate and local. Without visible, tangible benefits flowing back to the communities most affected, public resistance will only intensify.
Why the "Do Nothing" Approach Is No Longer an Option
There is a political temptation to delay action on AI-related economic disruption — to wait for clearer data, study the problem longer, or hope the market self-corrects. Senator Warner appears to have concluded that window has closed.
The signs are already in place. Workers are being displaced. Communities are angry. Public trust in AI is eroding. And the companies driving this transformation have largely avoided accountability for the economic disruption they are creating. A data center tax is not a perfect solution, and Warner himself has not yet committed it to paper as formal legislation. But it represents something increasingly rare in technology policy: a concrete, enforceable mechanism that ties industry growth to community benefit.
The AI boom is not going to slow down. Data centers will continue to be built. The question is not whether communities will bear the cost — they already are. The question is whether the industry will be required to share in the burden, or whether workers and local governments will be left to absorb the damage alone.
As Senator Warner put it, without that tangible benefit flowing back to communities: "The pitchforks are coming out."
That may be the most honest assessment yet of where this is heading — and why the conversation about who pays for the AI transition can no longer wait.
