Who Pays for the Power? The AI Boom Is About to Test the Electric Grid
Artificial intelligence has quickly become one of the most powerful forces shaping the global technology economy. From generative AI tools to advanced automation systems, companies across industries are racing to deploy new AI capabilities. Behind that race lies a massive and often overlooked constraint: electricity.
Training and operating modern AI systems requires enormous computing power, and that computing power consumes vast amounts of energy. As hyperscale data centers multiply across the United States and around the world, electricity demand is rising at a pace not seen in decades. The challenge now facing policymakers, utilities and technology companies is not simply how to produce more electricity. It is determining who should bear the cost of building the infrastructure required to deliver it.
This debate moved into the spotlight recently with the introduction of the Ratepayer Protection Pledge by the White House, an initiative aimed at ensuring that the rapid growth of AI and data center infrastructure does not shift the financial burden of new energy investments onto households and small businesses. The central principle behind the pledge is simple: growth should pay for growth.
So what does all of this have to do with Digital Infrastructure and Real Estate? Plenty, according to the people who ought to know. John Gilbert, former COO of Rudin management in New York City speaking at the recent Winter Sunshine AI, Data and Connectivity Summit said, “When Jack Rudin thought about building a new building the first person he called was the Head of Con Edison, our local utility. As we talk about Artificial Intelligence, electricity is a huge issue. At lunch with a dear friend who runs one of the largest utilities in the country, he mentioned that he had applications to create 29 gigawatts of power for datacenters. His current peak load is 13”. So yes, it matters for Real Estate, bigtime.
Under the framework being discussed, major technology companies expanding AI infrastructure would be expected to help fund the generation capacity and grid upgrades required to support their electricity demand. The pledge outlines several core expectations for large energy users, including bringing new power supply online, financing the grid infrastructure required to connect those facilities and structuring energy contracts so that existing utility customers are not exposed to additional risk.
The logic behind the proposal reflects a long-standing principle in infrastructure development. Historically, when communities invested in new energy infrastructure, local economies benefited directly through job creation, business growth and improved services. Power plants, factories and industrial facilities created economic activity within the regions where energy systems expanded.
AI infrastructure, however, changes that dynamic. A single hyperscale data center can consume as much electricity as tens of thousands of homes while serving users distributed across the country or even globally. The economic benefits of that infrastructure may not always flow to the communities where the electricity is produced or transmitted. This shift raises questions about how energy costs should be distributed and whether existing utility customers should subsidize infrastructure built primarily for large technology firms.
According to reporting from Reuters, the White House’s Ratepayer Protection Pledge is designed to ensure that the cost of powering data center expansion does not translate into higher electricity bills for ordinary consumers https://www.reuters.com/technology/white-house-seeks-protect-ratepayers-amid-data-center-power-boom-2025-09-19. The initiative reflects growing concern among regulators that the accelerating demand from AI facilities could put upward pressure on energy prices if infrastructure investments are not structured carefully.
Yet the most immediate challenge may not be electricity generation itself. Across the United States, the real bottleneck increasingly lies in access to the electric grid.
Developers seeking to connect new power generation or large electricity loads must navigate a complex and often slow interconnection process. In many regions, projects wait years for approval while utilities study the impact of new demand on transmission systems. These delays are compounded by limited transparency around available grid capacity, uncertainty about the cost of required upgrades and lengthy permitting processes for new infrastructure.
If managed correctly, however, the surge in electricity demand could catalyze one of the largest infrastructure investment cycles in modern history. Expanding the power system to support AI and electrification will require new generation resources, stronger transmission networks and modernized grid management technologies. These investments could create economic opportunities in host communities while strengthening the resilience and flexibility of the energy system.
The question ultimately returns to fairness. If large technology firms are driving unprecedented demand for electricity, many policymakers argue they should also play a leading role in financing the infrastructure required to supply it. Doing so could accelerate investment while protecting households and small businesses from higher energy costs.
As artificial intelligence continues to reshape industries, the energy systems that support it must evolve as well. The decisions being made today about cost allocation, infrastructure development and grid access will influence not only the future of AI but the stability of the electricity system itself. Ensuring that growth pays for growth may prove to be one of the defining policy challenges of the AI era.
For more information on the White House initiative aimed at protecting electricity customers from the costs associated with AI-driven data center growth, you can read the original article from Reuters at https://www.reuters.com/technology/white-house-seeks-protect-ratepayers-amid-data-center-power-boom-2025-09-19
