With Great power
Electricity Is Becoming Unbelievably Expensive as the US Power Grid Decays Into Ruin
Shocking.
https://futurism.com/electricity-usa-expensive-decaying?fbclid=IwdGRjcAUjgYFjbGNrBSOBamV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeg_f2vEPY-gQ8u561UnUmyS8IPPf9xYOB887qtEKi2ubOZDVODzpw4-hSRdQ_aem_SbnBMn5m_7J2c3tPPipmjQ

The US electrical grid is facing a stress test like never before. Thanks to a perfect storm of AI power consumption, climate crisis, crony capitalism, and a president bent on uprooting perfectly good energy infrastructure, the country’s already-struggling power system is rapidly crumbling as costs balloon into the stratosphere.

A recent analysis by Bloomberg underscores just how dire it’s getting. In the country’s largest continuous grid — a 13-state monstrosity managed by PJM Interconnection LLC — the media company notes that power costs have set record highs for two years in a row.

NVIDIA CEO Jensen Huang has pushed back against claims that the rapid expansion of AI data centers is automatically harmful to communities. His argument is that the enormous investment flowing into AI infrastructure can help finance new power generation, grid upgrades, and other infrastructure that may serve communities beyond the data centers themselves.

There is some evidence supporting part of that argument. Major technology companies are increasingly paying for power and infrastructure connected to their facilities. Meta, for example, says it pays for the energy and water used by its data centers and has invested hundreds of millions of dollars in water and wastewater infrastructure. These investments can provide broader infrastructure benefits when they are structured properly.

Huang has also highlighted changes in data-center cooling. New AI facilities can use closed-loop liquid-cooling systems that circulate the same coolant repeatedly instead of constantly consuming fresh water for cooling. Meta says most of its newest AI-optimized facilities use this approach, while other new projects are also adopting low-water or air-assisted cooling designs.

But saying that AI data centers have little or no local environmental impact would go too far. Their electricity demand is substantial, and the effect can vary dramatically depending on the location, power system, cooling technology, and available water resources. During the 2026 U.S. heatwave, utilities and regulators raised concerns that data-center growth was advancing faster than some electricity and water infrastructure could expand.

Public concern is also measurable. A September 2026 survey of 3,424 U.S. adults found that 63% were very or extremely concerned that data centers could increase electricity prices, while 57% expressed similar concern about local water supplies. The same survey found that 53% were highly concerned about AI’s overall environmental impact.

The debate is becoming important enough that governments are responding. California recently enacted legislation requiring greater oversight of data centers, including issues involving electricity costs and water use. The European Commission has also proposed rules requiring large European data centers to disclose energy and water-efficiency information.

The reality is more complicated than either AI data centers are destroying communities or AI data centers have no environmental cost. Some facilities are investing heavily in energy, water, and grid infrastructure, while others can place significant pressure on local systems. The outcome depends on how projects are designed, who pays for new infrastructure, where the electricity comes from, and how scarce local resources are managed.

AI infrastructure is growing rapidly. The real question is whether that growth can deliver economic and technological benefits without shifting its costs onto the communities that host it.

Modern society increasingly depends on data centers. They power cloud computing, digital services, AI systems, online communication and countless businesses.

But unlike clean water, data centers are not a biological necessity for human survival.

That distinction becomes important as electricity and water demand from rapidly expanding digital infrastructure receives greater attention. Some data centers consume substantial amounts of water directly for cooling, while others use air cooling, closed-loop systems or alternative technologies.

At the same time, communities around the world face water scarcity, pollution and increasing pressure on freshwater resources.

The question is therefore larger than technology itself: how should societies balance digital infrastructure growth with essential environmental resources?

The answer isn’t necessarily to stop building data centers. It is to consider where they are located, how efficiently they use water and electricity, and whether alternative cooling and water-reuse systems can reduce their environmental footprint.

Technology serves society—but essential resources ultimately sustain it.

#WaterSecurity #DataCenters #AIInfrastructure #Sustainability #CleanWater