0001 AI Powerplants:
Hidden Engines for the New Machine Age

Welcome to the first edition of Technology In Our Lives.Thank you for being here at the very beginning.

My goal with this newsletter is straightforward: to take a thoughtful, grounded look at the technologies shaping our world — and to understand not just what they are, but how they shape the way we live, work, think, and interact.

Technology envelopes us.. It influences our decisions, routines, expectations, and even our sense of what is “normal.” But, so much of it operates out of sight, locked inside sealed buildings or embedded in systems, that we rarely think about it.

That’s why I write this: to make the unseen visible.

I’m R.A. Murphy — a retired tech professional and educator. I’ve spent more than five decades watching technology evolve, surprise, and sometimes overwhelm us. Over those years, I’ve learned that to understand the world we’re living in, we need to understand the systems driving it.

And today, one of the biggest and least understood of those systems is AI infrastructure — the physical machinery that makes artificial intelligence feel like magic.

A Note About How I Write

I work closely with AI tools when I write — not to replace my voice, but to support it. I guide the arguments, shape the tone, decide the structure, and choose the emphasis. The tools help me think more clearly and work more efficiently. They offer possibilities, but the meaning is mine.

I mention this because it reflects where we are now: we are living in a time of human-machine partnership. Understanding that partnership — its strengths, its limits, and its consequences — is central to understanding our relationship with modern technology.

INTRODUCTION — The Magic We Don’t See

Artificial intelligence seems wizard-like. You type a question and get an instant answer. You ask for an image and it appears. You consult your phone to find a nearby a restaurant, and you immediately get three choices.

To the end user, everything is smooth, fast, and almost effortless.

But the effortless thing is rarely the real thing.

Behind every instant response is a physical world we almost never think about — a world full of humming machinery, heat, electricity, cooling systems, water pumps, transformers, and fiber-optic networks.

AI doesn’t run on magic.
It runs on massive industrial infrastructure.

Artificial intelligence has become so good at hiding its machinery that it’s easy to forget it exists. But if we want to understand the world taking shape around us — economically, environmentally, politically — we need to understand the physical footprint behind the magic.

Let’s pull back the curtain.

SECTION 1 — What an AI Data Center Really Is

Most of us imagine “the cloud” as something light and airy, as if our data is floating above our heads among the clouds. It’s a poetic idea — but wildly inaccurate.

An AI data center is not a cloud. It is a factory.

A modern AI data center can stretch across dozens of acres and contain hundreds of thousands of high-performance computer chips packed tightly together. These chips — GPUs and specialized accelerators — generate extraordinary heat. To keep them alive, entire ecosystems of cooling systems, air-handling units, heat exchangers, and industrial chillers work nonstop.

If you stood inside one of these facilities (and you wouldn’t — security is tight), you’d feel an unbroken wall of sound: a deep, layered vibration from thousands of machines, fans, and compressors working at the edge of physical limits.

To call these buildings “warehouses” is to miss the point. They are power plants that consume, transform, and expel enormous amounts of energy.

And as AI grows more capable, these centers must grow larger, hungrier, and more complex.

We’re used to thinking of digital technology as clean and immaterial. But AI is pushing us into a new era, an era where the digital world requires physical resources at a scale few people have imagined.

SECTION 2 — The Energy Surge No One Expected

Here’s what catches almost everyone off guard:

A single large AI data center can use as much electricity as a U.S. city of 100,000 people.

Not a town.
Not a suburb.
A city.

For context, the average American household uses around ten thousand kilowatt-hours of electricity per year. Training a large AI model — just the training stage — can use several million kilowatt-hours. And after training, running the model daily consumes energy continuously.

This sudden demand has created a ripple effect:

Utility companies are revising their long-term projections.

Cities and counties are renegotiating energy access.

Power grids are under pressure to expand capacity faster than planned.

Lawmakers are scrambling to understand the implications.

None of this was on the radar ten years ago. Even five years ago, most analysts underestimated the energy appetite of AI by a wide margin.

One reason is that AI systems scale not linearly, but exponentially. To build smarter models, you need more data, more computation, and more power — at increasing levels.

We are witnessing the early days of a technological era that demands infrastructure at near-industrial levels, and this shift is happening much faster than the public conversation suggests.

SECTION 3 — The Other Half of the Story: Water

Electricity is only half the story.

AI also runs on water — often hundreds of thousands of gallons per day.

Why water?
Because all those high-performance chips produce immense heat, and many data centers use water-based cooling systems to keep temperatures stable.

Even centers that use air cooling systems often rely on water behind the scenes for chillers and evaporative cooling.

This raises uncomfortable questions, especially in water-stressed regions:

Should AI companies receive priority access during droughts?

How should cities manage water allocation between community use and industrial technology use?

What happens when multiple centers cluster in the same geographic area?

Some companies now publish water usage reports; others do not. Transparency varies widely.

It’s not about “AI is bad” — that’s far too simplistic. It’s about AI has physical consequences, and we’re only now starting to grapple with them as these facilities multiply.

SECTION 4 — The Nuclear Shift

There’s a growing realization inside the tech world:

If AI keeps expanding at this pace, the existing grid won’t be able to support it.

This is not speculation — it’s already happening.

And tech companies are responding.

The phrase you’ll be hearing a lot more in the coming years is SMR — Small Modular Reactor.

These are next-generation nuclear reactors designed to be:

compact

factory-built

scalable

safe-demand designs

capable of powering data centers directly

To be clear, this does not mean every data center will have a nuclear reactor next door. But the trend is unmistakable:

Tech companies want dedicated, reliable, always-on power. Nuclear provides that. Several major companies are already exploring partnerships with SMR developers.

If AI is the engine of the new machine age, nuclear may become its fuel.

This shift will affect public policy, environmental debates, job creation, and even energy prices. It’s a rare intersection where technology and national infrastructure collide directly.

SECTION 5 — A Quiet Land Rush

Data centers are not small buildings you can tuck behind a mall. They require:

huge parcels of land

high-voltage grid access

water availability

fiber connectivity

specific zoning

geological stability

Because of this, regions across the U.S. are experiencing what some analysts are calling a data center land rush.

Farmland, industrial zones, and rural properties are being surveyed and purchased at rates unseen in decades. Local governments are being approached with proposals for billion-dollar facilities that promise tax revenue and construction jobs, but also bring concerns about noise, water, and long-term environmental impact.

Some communities welcome these projects. Others push back. Most are still learning what a commitment like this means.

But because the pace is so fast, many regions feel pressured to make decisions before the full implications are understood.

It’s a moment of both opportunity and uncertainty — and one that will shape the physical geography of parts of the country for years to come.

SECTION 6 — Should We Be Concerned?

Concern is reasonable. Worry is not.

We are in the middle of a major technological transition — one comparable to the early days of electrification or the rise of the interstate highway system. These shifts always bring complications, and they always require time to understand and adjust.

Some of the big questions include:

How do we balance AI expansion with environmental responsibility?

How quickly should we build?

Which regions should host these facilities?

How do we ensure energy prices remain stable?

What role should government play in regulating or supporting growth?

There are no simple answers.

But there is good news: engineers, policymakers, utilities, and environmental scientists are already deeply involved in these conversations.

The key going forward is public awareness.

We don’t need panic.

We need understanding.

When we understand the infrastructure behind AI, we can make informed decisions about how we want these systems to shape our future.

SECTION 7 — What This Means for Most People

So how does all of this affect everyday life?

Here’s what may happen over time:

1. Rising Energy Costs in Some Regions

Electricity prices may fluctuate as utilities spend to upgrade infrastructure.

2. More Data Centers Being Built

Construction will become visible — especially in rural states and counties with land availability.

3. New Debates Over Water Use

Communities with limited water resources will face difficult choices.

4. Subscription Model Changes

As operating costs increase, AI companies may adjust pricing for premium features.

5. Job Growth and New Skills

The rise of AI infrastructure will create demand for technicians, electricians, IT specialists, and logistics workers.

6. Faster Adoption of Nuclear and Renewable Power

AI expansion may accelerate the shift toward carbon-free energy.

7. A New Public Understanding of Digital Reality

More people will begin to realize that “the cloud” isn’t a cloud — it’s a network of very real, very physical machines.

CONCLUSION — Seeing the World Through Technology

One of the central messages of this newsletter is that technology is not abstract. It’s physical. It’s material. It has weight and heat and cost and consequence.

AI is no exception.

The systems that feel magical on our screens are powered by massive industrial engines humming at the edges of our communities. These engines shape everything from energy policies to land use to water rights — and soon, nuclear strategy.

Understanding this doesn’t require fear.

It requires perspective.

The more we understand the machinery behind our modern tools, the better prepared we are to navigate the world they create — thoughtfully, responsibly, and with clear vision.

Prefer listening to reading?

I record full audio editions of my essays so you can listen on your own schedule.
You’ll find the audio player at the top of each post.

— R.A. Murphy
TechnologyInOurLives.coms is a Paragraph Font