{"id":123088,"date":"2026-02-25T13:58:08","date_gmt":"2026-02-25T13:58:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/us-pa\/123088\/"},"modified":"2026-02-25T13:58:08","modified_gmt":"2026-02-25T13:58:08","slug":"ais-growing-appetite-for-power-is-putting-pennsylvanias-aging-electricity-grid-to-the-test","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us-pa\/123088\/","title":{"rendered":"AI\u2019s growing appetite for power is putting Pennsylvania\u2019s aging electricity grid to the test"},"content":{"rendered":"

The rapid growth of data centers<\/a> that support artificial intelligence is reshaping how electricity systems operate across the United States. <\/p>\n

Pennsylvania is emerging as a key location<\/a> in this shift, particularly around Pittsburgh. Access to existing infrastructure \u2013 such as repurposed industrial sites \u2013 the availability of skilled labor and a dense population make the region attractive to developers. <\/p>\n

Pittsburgh\u2019s large, dense population creates strong demand for fast, local computing power. The city\u2019s existing industrial, energy and academic resources make it easier and cheaper to build out that infrastructure quickly.<\/p>\n

As an assistant professor at the Heinz College of Information Systems and Public Policy at Carnegie Mellon University, I\u2019ve spent much of my career researching<\/a> machine learning, data science and optimization. The rise of data centers in the Pittsburgh region raises a key question for me: Can the electric grid can meet the demand of these facilities without increasing risks for everyone who depends on it?<\/p>\n

Why data centers stress the grid differently<\/p>\n

Unlike most commercial or industrial customers, large data centers consume enormous amounts of power<\/a> around the clock. A single modern facility can consume as much electricity as tens of thousands of homes<\/a>. <\/p>\n

AI-focused data centers are even more energy-intensive because they rely on high-performance computers that run continuously. Other data centers<\/a> focus on tasks such as hosting websites, storing files or managing databases. Unlike AI-focused data centers, these workloads fluctuate in demand and allow hardware to enter low-power states during quiet periods. <\/p>\n

The challenge for grid operators isn\u2019t just the amount of electricity but the immediate need for it. Utilities typically plan new generation, transmission and substations years in advance. Data center developers often expect power connections within months.<\/p>\n

Across the country<\/a>, this mismatch has created tension. Utilities in several regions have delayed or rejected interconnection requests because they cannot guarantee reliable service without major upgrades. <\/p>\n

Pennsylvania\u2019s grid faces similar pressures. Much of the state\u2019s electric infrastructure was designed for slower, more predictable growth. When multiple, large data centers cluster in a small area like greater Pittsburgh, they can overwhelm local substations and transmission lines<\/a>. This increases the risk of outages for all those connected to the grid or forces utilities to make costly emergency upgrades.<\/p>\n

Reliability is not just about keeping the lights on<\/p>\n

Reliability problems are often portrayed as major, systemwide blackouts that occur only occasionally but leave thousands of customers without power. In reality, grid stress shows up in more subtle ways: voltage fluctuations, equipment overheating and longer recovery times after storms or heat waves.<\/p>\n

\"High<\/a><\/p>\n

Local electrical grid substations and transmission lines can be overwhelmed when large data centers cluster in the same area.
\n Joe Raedle\/Getty Images News Collection<\/a><\/p>\n

These risks matter more in regions like Pittsburgh, where extreme weather is becoming more frequent<\/a> and the electrical infrastructure is aging<\/a>. A grid that is already strained has less margin to absorb sudden increases in demand.<\/p>\n

In our research on power grid resilience<\/a>, we have seen how meeting the demand for multiple data centers in the same area can amplify disruption when something goes wrong.<\/p>\n

In Indianapolis<\/a>, for example, neighborhoods served by the same grid often experienced outages during major weather events. When an increased demand for power is concentrated in one area, a single equipment failure can trigger a domino effect. <\/p>\n

Pittsburgh and other Pennsylvania cities share many of the same structural characteristics: high electrical or utility demand, older substations and limited space for expansion of the power grid. Adding multiple data centers into this mix raises the stakes.<\/p>\n

Who pays when upgrades are needed?<\/p>\n

One of the hardest questions utilities and regulators face is how to allocate the cost<\/a> of grid upgrades.<\/p>\n

Utilities traditionally spread infrastructure costs<\/a> across all customers. But when upgrades are driven by a few large customers, that approach<\/a> becomes controversial. Residential customers may pay higher bills<\/a> to support data center infrastructure that doesn\u2019t benefit them directly.<\/p>\n

Some states, including Virginia, Oregon, Ohio, Minnesota and Illinois, are experimenting with money-saving options, such as requiring data center developers to pay more up front for infrastructure<\/a>. Others, including Pennsylvania<\/a> and Missouri<\/a>, are exploring special tariffs on large power users \u2013 such as data centers \u2013 that reflect the true reliability and capacity costs of serving these facilities.<\/p>\n

\"Construction<\/a><\/p>\n

The rapid rise of AI has turned data centers \u2013 like this 480,000-square-foot center under construction in Ohio \u2013 into one of the fastest-growing sources of electricity demand in the country.
\n The Washington Post Collection\/Getty Images<\/a><\/p>\n

Pennsylvania is at an early stage in this debate. Decisions made now will shape not only electricity prices but also where data centers choose to locate. This also impacts how resilient the grid remains over the next decade.<\/p>\n

Planning for growth without sacrificing resilience<\/p>\n

Data centers can bring investment, tax revenue and jobs<\/a> to Pennsylvania. Our resilience research shows<\/a> that proactive planning can make them less disruptive. Utilities that wait until interconnection requests arrive often face higher costs and greater reliability risks, our research shows. Utilities that plan for future growth and build extra backups into their systems are much better at handling sudden hits \u2013 such as heat waves, equipment failures or massive surges in power use. <\/p>\n

Advanced forecasting tools<\/a> can help. By combining historical outage data, weather patterns and projected load growth, planners can identify which substations and corridors are most likely to become bottlenecks. This approach allows targeted upgrades rather than broad, expensive overhauls.<\/p>\n

\"A<\/a><\/p>\n

The Bruce Mansfield Power Plant in Shippingport, Pa., is likely to become the next AI-powered data center in the area.
\n Robert Nickelsberg\/Getty Images News Collection<\/a><\/p>\n

Another option is encouraging data centers to be more flexible in how they consume power. Some data center operators<\/a> are beginning to invest in on-site generation, battery storage or shifting noncritical computing tasks to off-peak hours. While these measures cannot eliminate grid impacts, they can buy valuable time and reduce risk. <\/p>\n

A local issue with national implications<\/p>\n

What happens in Pittsburgh will not stay in Pittsburgh. Pennsylvania sits at the center of a regional grid<\/a> that serves much of the mid-Atlantic and Midwest. Reliability challenges in one area can ripple outward, especially during extreme events.<\/p>\n

The rapid rise of AI has turned data centers into one of the fastest-growing sources of electricity demand in the country. How states like Pennsylvania respond will influence national conversations about grid modernization, cost allocation and resilience.<\/p>\n

If planners treat data centers as just another large customer, the grid may struggle to keep up. If they recognize that this new wave of demand changes the rules of reliability, there is an opportunity to strengthen the system for everyone.<\/p>\n","protected":false},"excerpt":{"rendered":"The rapid growth of data centers that support artificial intelligence is reshaping how electricity systems operate across the…\n","protected":false},"author":2,"featured_media":123089,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[28,30,29],"class_list":{"0":"post-123088","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-pennsylvania","8":"tag-pennsylvania","9":"tag-pennsylvania-headlines","10":"tag-pennsylvania-news"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts\/123088","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/comments?post=123088"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/posts\/123088\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/media\/123089"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/media?parent=123088"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/categories?post=123088"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us-pa\/wp-json\/wp\/v2\/tags?post=123088"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}