White House expands commitments on data center power procurement and grid integration to meet AI infrastructure demand
The White House has expanded a voluntary initiative aimed at preventing data center infrastructure costs from being transferred to household and business electricity customers. On July 23, the administration announced that more than 200 utilities, electric cooperatives, data center developers, technology companies, and state governments have joined its Ratepayer Protection Pledge. According to the administration, these participants collectively represent approximately 80% of electricity delivered to U.S. homes and businesses and cover areas where 263 million people live. These figures have not been independently audited, and the pledge itself carries no binding force as federal rate order, utility tariff, or infrastructure contract; its practical effect will depend entirely on how its commitments are translated into enforceable agreements and regulatory decisions.
The pledge framework contains five core commitments. Its central principle holds that new large-load customers should pay the generation and grid costs their projects create rather than spreading those expenses across the wider customer base. The White House cited utility agreements in Michigan, Indiana, Georgia, Mississippi, Wisconsin, Louisiana, and Texas as examples of data center developers assuming infrastructure costs or generating projected customer benefits—though these savings estimates rest on administration-cited figures and may depend on future energy use, construction costs, market prices, and regulatory treatment.
Artificial intelligence computing, cloud services, streaming, financial systems, and other digital services are driving construction of massive facilities requiring hundreds of megawatts of continuous electricity. In some markets, a single data center campus can create demand comparable to an entire city or major industrial complex. Meeting that load often requires new generation capacity, substations, transmission lines, distribution upgrades, capacity purchases, and backup systems. Disputes frequently arise when utilities make long-term investments for projects whose construction schedule, electricity use, or operating life later changes.
This issue emerges as retail electricity costs are already rising nationwide. The U.S. Energy Information Administration reported that average residential revenue per kilowatt-hour reached 18.44 cents in May 2026—6.2% higher than May 2025. PJM Interconnection, which coordinates the power grid across all or parts of 13 states and the District of Columbia, saw data center demand account for nearly 40% of its latest capacity-market charges, making these costs relevant to a large share of the eastern United States.
The pledge remains voluntary and establishes no uniform national enforcement mechanism. While it brings utilities, developers, technology companies, cooperatives, and state governors into a shared policy framework, participation alone does not determine how costs will be allocated in any individual utility territory. Actual protections typically must appear in large-load tariffs, minimum-payment provisions, construction agreements, interconnection contracts, regulatory orders, or other binding arrangements—details that may vary significantly by state, utility, market structure, and specific project.
Practical protections can take multiple forms: a separate rate class requiring data centers to pay for reserved capacity even below projected usage; construction agreements assigning responsibility for new substations or transmission connections; or financial security requirements protecting other customers if a project is delayed or canceled. Without such provisions, a broad public commitment may offer limited recourse if costs are ultimately shifted to other ratepayers.
The administration has published lists of participating organizations, but an organization's presence on the pledge does not itself indicate which obligations are legally enforceable or which specific projects they cover. It remains unclear whether the White House's national coverage calculation measures utility service territories, electricity sales, participating companies, state populations, or some combination thereof. While the administration describes 263 million Americans as covered when a data center is built nearby, the actual degree of protection depends entirely on applicable utility rules and contracts.
The pledge also lacks a single national definition of which generation or grid investments are "caused" by a data center. Cost-allocation disputes frequently involve facilities serving both a large customer and the wider grid, complicating the division of costs beyond a single project invoice.
Key provisions will likely include minimum monthly payments, contract duration, collateral requirements, responsibility for canceled projects, allocation of transmission and generation costs, treatment of unused capacity, and protections if a data center's electricity demand falls below its original forecast. Regulators may also examine whether new generation improves reliability for other customers, whether backup resources can operate during grid emergencies, and whether infrastructure built for a data center creates lasting benefits after the original contract expires.
The pledge establishes a public policy principle: large data centers should pay the costs associated with serving their demand. Whether that principle materially limits customer bills will ultimately be determined through the contracts and regulatory decisions that follow.