For a long time, Americans took electricity for granted. We flip a switch, charge a cell phone, watch a movie, or ask digital assistants what to cook for dinner. But behind this apparent ease lies enormous and growing pressure on the country’s electrical infrastructure. As artificial intelligence tools expand and data centers multiply, the U.S. power grid is beginning to show signs of stress, and the consequences are already reaching homes in the form of higher bills, overloaded appliances, and a greater risk of blackouts.
AI’s original dream resurfaces
What’s behind the so-called “1956 prophecy”? It goes back to the Dartmouth Workshop, when artificial intelligence was first imagined as a real discipline. Researchers dreamed of machines that could “think.” What they didn’t picture was that, seven decades later, those machines would guzzle electricity at a scale that rivals entire nations.
Today, that dream is alive inside sprawling data centers that run everything from chatbots to TikTok algorithms. A recent study from Lawrence Berkeley Lab puts the numbers into perspective: between 2014 and 2018, data centers were already adding around 7% more power use each year. From 2018 to 2023, the increase jumped closer to 20%. If the trend continues, by 2028 almost one in every ten U.S. kilowatts could be feeding AI alone. And the pain isn’t abstract. In Connecticut, energy rates are up more than 18%. In Maine, the jump is over 36%. Why? Because these models run nonstop, chewing through data at a massive scale. Newsweek summed it up bluntly:
“To keep pace, utilities are increasingly relying on aging fossil fuel plants to generate enough electricity to meet the crushing demand.”
America’s grid under siege
The consequences of this increase are real and palpable. It’s no wonder that PJM Interconnection, the largest regional power transmission organization in the US, is facing a critical shortage, unable to support new data center projects. Recent reports put regional electricity costs at $16.1 billion, highlighting the severe strain on the grid.
Industry experts, such as Joe Bowring, president of Monitoring Analytics, recommend that operators consider self-sufficient energy solutions: building in-house power generation in data centers can help them survive without further destabilizing the grid. These measures also serve as a sustainability test, benefiting projects that demonstrate financial and technological resilience. The imbalance between supply and demand has broader consequences:
- Record prices in recent auctions reveal a market that favors suppliers while burdening ordinary consumers.
- Backup generators in data centers, designed to protect the facility itself, can overload the local grid, causing near-cascading blackouts, leading experts to warn that the internet is about to collapse in America. Reuters noted that this “has forced operators to plan for new contingencies, complicating the already difficult task of balancing electricity supply and demand in the country.”
- Living near these data centers shorten the lifespan of appliances, increase the risk of overheating, and even trigger electrical fires.
AI’s hidden bill
The big statistics grab headlines, but the real weight lands at home. Parents staring at swollen utility bills. Kids losing sleep when lights flicker. Washing machines dying years before their time. Neighbors frustrated by the glow and heat pouring out of massive facilities that were supposed to represent “the future.” Put simply:
- Households are footing costs that were never part of the AI sales pitch.
- Communities are stuck with the noise, pollution, and stress of mega-projects.
- Cities are scrambling to keep growth from wrecking daily life.
The rapid adoption of AI tools has outpaced the infrastructure’s ability to support it. Without careful planning and incentives for energy self-sufficiency, the financial and stability impacts of the grid will only worsen. That’s why new technologies are already being considered, an example of which is this invention unveiled by America, which performs a quintillion operations per second.
Disclaimer: Our coverage of events affecting companies is purely informative and descriptive. Under no circumstances does it seek to promote an opinion or create a trend, nor can it be taken as investment advice or a recommendation of any kind.
