Households across the mid-Atlantic are facing higher electricity bills tied to transmission upgrades they did not request, built to serve AI-driven data centers they will never visit. Maryland’s Office of People’s Counsel has challenged PJM cost rules that would assign roughly $2 billion in data-center-driven transmission costs to residential and small-business ratepayers. At the federal level, the Department of Energy and the Federal Energy Regulatory Commission have both moved to address the surge in large-load electricity demand, but the central question of who pays for grid expansion remains unresolved.
How data-center demand growth hits household electric bills
The mechanism is straightforward. When a hyperscale data center connects to the grid, it often triggers transmission upgrades: new high-voltage lines, substation expansions, and transformer installations. Under current rules in many regions, those costs are spread across all ratepayers in the same grid territory, regardless of whether a household or small business benefits from the new capacity. The result is that a retiree in Baltimore or a restaurant owner in Richmond can see a line-item increase on a monthly bill driven by infrastructure built for a facility hundreds of miles away.
The Energy Department recently highlighted how rapidly expanding data-center load is reshaping grid planning. Drawing on analysis from Lawrence Berkeley National Laboratory, the report describes AI-heavy facilities that use far more power per rack than traditional cloud computing, compressing years of expected demand growth into a much shorter window. That acceleration forces transmission planners to move faster, locking in large capital projects whose costs are then recovered from customers over decades.
Because transmission is a shared network, regulators have historically treated most of those projects as regional assets, with costs spread broadly. That model made sense when new lines primarily supported general load growth or reliability for everyone. The new wave of demand, however, is highly concentrated in specific clusters of data centers, raising the question of whether the same cost-sharing logic still applies. Consumer advocates argue that when a single industry drives disproportionate upgrades, broad allocation risks turning household bills into a quiet subsidy for corporate infrastructure.
FERC’s push to manage large-load integration
Federal regulators have begun to grapple with that shift. The Federal Energy Regulatory Commission, which oversees wholesale power markets and interstate transmission, has launched what it calls an aggressive, targeted effort to manage the influx of large new loads. Through a series of show-cause orders, FERC has directed regional grid operators to explain how they will integrate data centers and other big customers without undermining reliability or unfairly shifting costs to existing ratepayers.
Those orders focus on the nuts and bolts of interconnection queues, studies, and tariffs. FERC is pressing operators to clarify when a new customer must fund upgrades directly and when costs can be rolled into general transmission rates. The commission is also probing whether current rules allow some large users to structure projects in ways that minimize their own contributions while still relying heavily on the shared grid.
Nowhere is that concern more visible than in the commission’s directive to PJM on co-location. In a separate action, FERC instructed the country’s largest grid operator to develop clearer rules for arrangements in which data centers connect alongside or directly to power plants. The co-location guidance reflects a worry that data centers can partially bypass the network-reducing the transmission charges they pay-while still counting on it for backup and ancillary services. If tariffs do not fully capture that reliance, other customers may be left to shoulder more of the systemwide costs.
Maryland’s $2 billion cost fight with PJM
The sharpest illustration of the billing dispute is playing out in Maryland. The state’s Office of People’s Counsel has filed a complaint challenging PJM’s cost-allocation methodology, arguing it would saddle Maryland ratepayers with roughly $2 billion in transmission investments driven by data-center expansion. The complaint targets the way PJM assigns regional transmission charges, contending that the formulas do not reflect who actually causes the need for new infrastructure.
Northern Virginia, which sits within PJM’s footprint, hosts one of the densest concentrations of data centers in the world. When those facilities require grid upgrades that PJM classifies as regional rather than local, the costs flow outward to all load-serving entities in the region, including Maryland utilities and their customers. OPC’s position is that cost-causation principles should apply: the entities whose demand triggers the upgrade should bear the expense, not households and small businesses hundreds of miles away that see no direct benefit.
The stakes for Maryland consumers are concrete. Transmission costs are embedded in delivery charges on monthly bills, and a multibillion-dollar build-out can translate into persistent increases that show up for years. OPC argues that under PJM’s current rules, state regulators have limited ability to shield residents from charges tied to projects outside their borders and beyond their control. The case has become a test of whether regional planning can adapt to an era in which industrial clusters, rather than broad-based growth, drive much of the need for new lines.
Will federal reforms slow residential rate increases?
FERC’s recent actions suggest the commission recognizes that existing rules were not designed for the speed and concentration of AI-driven load growth. The show-cause orders amount to a demand for accountability: grid operators must justify their interconnection processes, explain their cost-allocation methods, and propose changes where current practice no longer fits. The directive to PJM on co-location goes further, forcing the operator to close gaps that allowed data centers to negotiate arrangements that lowered their own costs while potentially increasing costs for everyone else.
Whether these moves will materially slow residential rate increases is less clear. Transmission planning and approval cycles typically span many years, and disputes over who pays for what can take just as long to resolve through formal FERC proceedings. In the meantime, developers are signing power contracts and breaking ground on new facilities, betting that the grid will expand in time to meet their needs. Even if FERC ultimately tightens cost-causation rules, some of the most expensive projects now in the pipeline may already be locked into existing allocation frameworks.
Another challenge is that the impacts are uneven. Regions that embrace stricter cost-causation-requiring large new loads to fund a bigger share of the transmission they trigger-could see slower growth in residential rates than regions that continue to socialize those costs widely. But the comparison is complicated by differences in legacy infrastructure, fuel prices, and state policies. A state with abundant low-cost generation might absorb new data-center demand with relatively modest upgrades, while another with constrained capacity could face major reinforcements even for similar levels of growth.
Data gaps and what consumers can watch
Several gaps in the public record limit how precisely anyone can project the household bill impact. The DOE report references detailed LBNL modeling of data-center demand but does not provide granular, state-by-state estimates of how many cents per kilowatt-hour residential customers might pay as a result. FERC’s proceedings, meanwhile, contain technical discussions of tariff language and planning criteria, but few plain-language summaries that connect specific rule changes to monthly bills.
For now, consumers and local officials are left to track a few key signals. One is the volume of large-load interconnection requests in their region, which hints at how much new infrastructure may be required. Another is the outcome of cases like Maryland’s, which will help define how far cost-causation principles extend when a single industry drives disproportionate growth. Finally, FERC’s evolving guidance to grid operators will determine whether the next wave of data-center expansion proceeds under rules that more clearly distinguish between shared reliability investments and facilities built primarily for private benefit.
The underlying tension is unlikely to disappear. AI and cloud computing continue to expand, and the grid must grow with them. The policy question is whether that growth reinforces long-standing norms of broad cost-sharing or marks a shift toward expecting large, sophisticated customers to pay a greater share of the upgrades they require. For households across the mid-Atlantic and beyond, the answer will show up not in federal dockets, but in the quiet upward drift-or restraint-of the delivery charges on their electric bills.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
