Skip to Content
Climate change and energy

The Debrief: Power and energy

AI’s footprint today is likely the smallest it will ever be; how are we going to power it in the future?

June 25, 2025

It may sound bluntly obvious, but energy is power. Those who can produce it, especially lots of it, get to exert authority in all sorts of ways. It brings revenue and enables manufacturing, data processing, transportation, and military might. Energy resources are arguably a nation’s most important asset. Look at Russia, or Saudi Arabia, or China, or Canada, or Qatar, or—for that matter—the US. For all these nations, energy production plays key roles in their economies and their outsize global status. (Qatar, for example, has a population roughly the size of metro Portland, Oregon.) 

The US has always been a nation of energy and industry. It was a major producer of coal, which fed the Industrial Revolution. World War II was won in large part by the energy production in the United States—which fueled both manufacturing of the war machine at home and its ships, planes, and tanks in the Pacific and Europe. Throughout its history, the country has found strength in energy production. 

Yet in many ways right now the US seems to be forgetting those lessons. It is moving backward in terms of its clean-­energy strategy, especially when it comes to powering the grid, in ways that will affect the nation for decades to come—even as China and others are surging forward. And that retreat is taking place just as electricity demand and usage are growing again after being flat for nearly two decades. That growth, according to the US Energy Information Administration, is “coming from the commercial sector, which includes data centers, and the industrial sector, which includes manufacturing establishments.” 

As MIT Technology Review has extensively reported, energy demand from data centers is set to soar, not plateau, as AI inhales ever more electricity from the grid. As my colleagues James O’Donnell and Casey Crownhart reported, by 2028 the share of US electricity going to power data centers may triple. (For the full report, see technologyreview.com/energy-ai.)

Both manufacturing and data centers are obviously priorities for the US writ large and the Trump administration in particular. Given those priorities, it’s surprising to see the administration and Congress making moves that would both decrease our potential energy supply and increase demand by lowering efficiency. 

This will be most true for electricity generation. The administration’s proposed budget, still being considered as we went to press, would roll back tax credits for wind, solar, and other forms of clean energy. In households, they would hit credits for rooftop solar panels and residential energy efficiency programs. Simultaneously, the US is trying to roll back efficiency standards for household appliances. These standards are key to keeping consumer electricity prices down by decreasing demand. 

In short, what most analysts are expecting is more strain on the grid, which means prices will go up for everyone. Meanwhile, rollbacks to the Inflation Reduction Act and to credits for advanced manufacturing mean that fewer future-facing energy sources will be built. 

This is just belligerently shortsighted. 

That’s especially true because as the US takes steps to make energy less abundant and more expensive, China—our ostensible chief international antagonist—is moving in exactly the opposite direction. The country has made massive strides in renewable energy generation, hitting its goals six years ahead of schedule. In fact, China is now producing so much clean energy that its carbon dioxide emissions are declining as a result.

This issue is about power, in all its forms. Yet whether you’re talking about the ability to act or the act of providing electricity, power comes from energy. So when it comes to energy, we need “ands,” not “ors.” We need nuclear and solar and wind and hydropower and hydrogen and geothermal and batteries on the grid. And we need efficiency. And yes, we even need oil and gas in the mid term while we ramp up cleaner sources. That is the way to maintain and increase our prosperity, and the only way we can possibly head off some of the worst consequences of climate change.

Deep Dive

Climate change and energy

We did the math on AI’s energy footprint. Here’s the story you haven’t heard.

The emissions from individual AI text, image, and video queries seem small—until you add up what the industry isn’t tracking and consider where it’s heading next.

The data center boom in the desert

The AI race is transforming northwestern Nevada into one of the world's largest data-center markets—and sparking fears of water strains in the nation’s driest state.

Everything you need to know about estimating AI’s energy and emissions burden

Here’s how MIT Technology Review waded through a mess of data and hidden variables to calculate the individual and collective energy demand from AI.

Can nuclear power really fuel the rise of AI?

Tech giants are looking for more energy, but building new reactors takes time.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.