AI Energy Demand
Rapid AI Data Center Expansion Sparks Community Backlash and Energy Concerns
Anonymous
Key Points
1
More than 500 U.S. data-center projects have reportedly been delayed, primarily because local residents oppose having them nearby; estimates also vary because research groups use different counting methods.
2
The World Economic Forum estimates global data-center investment could reach $7 trillion, supported by the substantial capital access of companies such as Microsoft, Alphabet, Amazon and Meta, as well as Wall Street financing.
3
The Harwood, North Dakota, project is being built as two 907,000-square-foot structures, with roughly 1,800 workers involved. Applied Digital expects part of the facility to become operational by the end of 2026 and the remainder in early 2027.
4
Applied Digital plans to spend $112 million on the electrical substation and transmission upgrades needed for the Harwood facility, which will draw power from Cass County Electric.
5
The security debate extends beyond physical infrastructure: an article on frontier AI argues that companies are discussing autonomous AI engineers and superhuman coding while users still face basic transparency questions such as, “Who is using my AI account?”
The United States is racing to build AI data centers, with more than 3,000 already operating and roughly 1,500 under construction, according to The Alpena News. A $3 billion project in North Dakota will consume 280 megawatts of electricity—enough power for a small city—but local residents are pushing back hard. Communities across the country worry about rising electricity bills, fossil-fuel use, and land consumption The Alpena News.
The tech boom is creating a tension between national AI ambitions and local power grids. Financial Times reports the US and China are now discussing AI security together, even as American companies like Microsoft and Alphabet race to build massive computing campuses. The World Economic Forum warns that without addressing community concerns and grid capacity, up to $1 trillion in infrastructure spending could be wasted.
North Dakota's $3 Billion Power Battle
Applied Digital is building two 907,000-square-foot AI facilities near Harwood, North Dakota. The project will draw 280 megawatts of power and cost $112 million just for electrical upgrades The Alpena News. The company expects part of the campus to run by late 2026, with full operation in early 2027.
But the deal has sparked fierce local opposition. Hundreds of residents showed up to town halls demanding answers about power costs and environmental damage The Alpena News. Nearby Fargo even tried to annex 800 acres containing the site to gain tax revenue—triggering a legal battle. Mayor Blake Hankey accused Fargo of trying to "steal this project."
Hundreds of Projects Face Neighborhood Resistance
The North Dakota conflict is not an outlier. Data Center Watch reports that 120 U.S. data center projects worth nearly $200 billion were delayed or blocked during the first half of 2026, mostly due to local opposition. Over 500 projects nationwide have reportedly stalled as residents demand accountability.
A Gallup poll found that 70% of Americans oppose new AI data centers in their neighborhoods, citing rising electricity costs, noise, and water use. Investment firms have grown skeptical too. Some companies paused IPOs because of uncertainty around data center returns and public backlash, slowing the infrastructure boom's momentum.
Tech Giants Compete as Energy Demands Skyrocket
Microsoft, Alphabet, Amazon, and Meta collectively committed over $700 billion toward AI infrastructure in 2026 Seeking Alpha. The World Economic Forum estimates global data center investment could reach $7 trillion by 2030. But that growth depends on whether companies can access affordable power and overcome neighborhood opposition.
The International Energy Agency projects data center electricity use will nearly double by 2030, reaching 950 terawatt-hours. Streamline Feed reports European startups are deploying liquid cooling and battery software to reduce strain on electric grids. Yet the US debate is less about efficiency and more about whether these campuses should be built at all.
Security Concerns Lag Behind Hardware Expansion
As companies discuss autonomous AI engineers and superhuman coding, basic account protection still lags. Frontier AI research raises a simple question users want answered: "Who is using my AI account?" Security transparency has become an afterthought compared to raw computing power and speed-to-market.
The debate also centers on US-China competition. Financial Times reports the two nations agreed to establish formal dialogue on AI security, proposing a mutual notification system about AI developments. Yet local communities remain focused on immediate concerns: soaring power bills and depleted farmland near their homes, not geopolitical strategy.
Sentiment
Mostly PositiveNEGATIVENEUTRALPOSITIVE
Publishers
88
Articles
58
Reach
146
A
AnonymousAuthor
