Across the United States, the rapid expansion of AI data centers is forcing major grid upgrades at a pace rarely seen before, but in many cases, the cost of building the new infrastructure needed to support them is not being paid by the companies driving the demand. Instead, it is spread across residential customers through higher rates. This practice is now under growing scrutiny from lawmakers, regulators, and grid operators who warn that households are being asked to subsidize private infrastructure built for corporate profit.

Why Are Your Electricity Bills Rising Because of AI Data Centers?

Hyperscale data centers, built to train and run artificial intelligence models, can consume as much electricity as entire towns. When one of these facilities connects to the existing grid, it triggers infrastructure upgrades: transformer upgrades, new transmission lines, and substation expansions. Those costs get socialized through the rate base, meaning every household and small business in the utility's territory pays a share of infrastructure they never asked for and will never directly use.

The numbers are staggering. Pennsylvania residential electricity rates rose 21.7% in 2025, while West Virginia utility bills are now exceeding the average monthly mortgage payment in some households. Maryland electric bills are set to rise this summer as PJM, the grid operator serving 65 million people, struggles to handle data center demand it was not built to handle.

The U.S. Energy Information Administration projects that electricity demand from data centers will double by 2030. This abstract projection translates into real consequences for families already struggling with energy costs. The issue has become so pressing that it is now threatening incumbent politicians in four competitive Pennsylvania congressional races, transforming what was once an energy policy debate into a kitchen table economic issue in swing districts across the country.

What Are States Doing to Stop Cost-Shifting to Residents?

Three states have already moved to require data centers to pay for their own infrastructure. On April 24, regulators in Wisconsin issued a ruling requiring data centers to cover the full cost of the power infrastructure they require, explicitly stating that utilities "must not shift those costs onto other customers." The decision, issued by the Wisconsin Public Service Commission, marked a shift from policy debate to enforceable regulation built around the beneficiary-pays principle.

Three

Similar legislation has already passed in states including Florida and Pennsylvania during the current session, suggesting the approach is spreading quickly as more states move to prevent residential customers from absorbing the cost of private infrastructure. Florida passed SB 484 this legislative session, while Pennsylvania passed HB 2606, both requiring data centers to fund the capital costs of any transmission or distribution upgrades needed to serve their connection.

However, the policy landscape remains contested. In Michigan, a $16 billion Oracle data center project secured financing backed by Blackstone and PIMCO, even as the project's power supply agreements are facing a legal challenge before the Michigan Court of Appeals. Michigan Attorney General Dana Nessel filed an appeal April 17, challenging the Michigan Public Service Commission's conditional approval of the DTE-Oracle contracts, arguing the contracts should have been subject to a contested case process that would have allowed outside parties to conduct legal discovery and formally scrutinize the agreements before approval.

How to Protect Ratepayers From Data Center Infrastructure Costs

• Require Full Cost Coverage: Data centers must fund the capital costs of any transmission or distribution upgrades needed to serve their connection, ensuring no costs are shifted to residential customers through higher rates.
• Post Performance Bonds: Require data centers to post a bond or performance guarantee against the demand they project, protecting ratepayers if demand fails to materialize or contracts expire early.
• Stress-Test Demand Projections: Regulators must require utilities to stress-test demand projections before approving infrastructure investment, preventing overbuilding based on inflated forecasts that may never materialize.

Dr. Mark McNees, a Florida State University Jim Moran College of Entrepreneurship professor and author of the policy brief "Who Pays for the AI Grid?", explained the stakes. "The storm is not coming. It is here, and it arrived faster than most policymakers expected," he stated, noting that the structural problem is straightforward: a data center connecting to the existing grid triggers infrastructure upgrades, and those costs get socialized through the rate base.

"The risk of being left unchecked is not just high electricity bills. It is stranded infrastructure. If a utility overbuilds to serve demand projections that later prove inflated, and data center demand shrinks because AI efficiency improves or contracts expire, those capital costs do not disappear. They sit in the rate base for decades, paid by ratepayers," McNees explained.

Dr. Mark McNees, Jim Moran College of Entrepreneurship, Florida State University

There is also a timing dimension that makes this moment critical. The grid infrastructure for AI data centers is being decided right now, in state legislatures and regulatory proceedings across the country. Unlike the Net Neutrality debate of the early 2000s, which centered on infrastructure that had already been paid for, the policy window for shaping AI data center costs is open and moving fast.

What Happens If Demand Projections Turn Out to Be Wrong?

Questions are being asked about whether projected data center demand will materialize at the scale utilities once expected. The Electric Reliability Council of Texas (ERCOT), which manages electricity for roughly 90% of the state, recently acknowledged that its long-term demand forecasts are "likely overstated," raising concerns that consumers could end up funding infrastructure built for growth that never fully arrives.

This risk is particularly acute because utilities and data center developers have strong financial incentives to overestimate demand. If a utility overbuilds to serve demand projections that later prove inflated, the capital costs do not disappear. They sit in the rate base for decades, paid by ratepayers who had no say in the decision. This is cost externalization at one remove, and current policy frameworks in most states do not require regulators to stress-test demand projections before approving the infrastructure investment.

The Oracle-linked Saline Township project in Michigan illustrates the scale of these investments. At full buildout, the facility is expected to draw roughly 1,383 megawatts of power, more than 10% of DTE's projected 2026 system peak of approximately 10.71 gigawatts. The project, known as "The Barn," comprises three single-story buildings totaling more than 1.6 million square feet and is targeted for completion in the fourth quarter of 2027.

The financing announcement by Related Digital stated the project is expected to generate approximately $300 million in savings for DTE ratepayers by contributing to fixed grid costs. However, the Michigan Attorney General's challenge suggests that the true cost allocation and ratepayer protections remain contested.

As more states grapple with this issue, the fundamental question remains: should households and small businesses subsidize the infrastructure costs of private companies building AI data centers? The answer, increasingly, is no. But the implementation of that principle, and the protection of ratepayers from stranded infrastructure costs, will require sustained regulatory vigilance and political will in the months and years ahead.