DARK SIDE OF THE SWOON

This story will make your blood boil. And it happened right here in America just weeks ago.

Residents in Fayette County, Georgia, were told by their local government to stop watering their lawns. Water pressure in their homes was dropping. People couldn’t shower properly. Something was draining the town dry.

So they started investigating. And what they found will shock you.

A massive data center campus located about 20 miles south of Atlanta had been secretly drawing approximately 29 million gallons of water through two water connections that the county didn’t even know existed — and this went on for 15 months before anyone caught it. 

15 MONTHS. Hidden pipes. Millions of gallons. While regular families were being told to stop watering their grass.

Now here’s the part that will really get you angry —

Officials refused to fine the builders of the massive 6.2 million-square-foot facility over the unauthorized water use. The county’s soft response was attributed to the data center being described as “our largest customer, and we have to be partners.” 

So let’s get this straight. A tech company secretly taps into your water supply without permission, drains 29 million gallons while your water pressure tanks, and the punishment is… nothing. Because they’re a “partner.”

This is happening all across America right now.

Data centers powering AI systems consumed approximately 17 billion gallons of water in 2023, and that number is projected to explode to 68 billion gallons by 2028 — a 300% increase in just five years. 

On average, a single medium-sized data center consumes roughly 110 million gallons of water per year — enough to cover the annual water use of about 1,000 households. 

And they’re building thousands of these things. Everywhere.

Many AI data centers are deliberately built in the country’s driest regions to take advantage of cheap solar power — placing enormous strain on water supplies in areas least equipped to handle the extra demand. 

Training a single large AI model like GPT-4 consumed approximately 700,000 liters of water for cooling. And generating just 10–50 AI responses consumes roughly 500 milliliters — about one water bottle. 

Scale that to billions of daily users across every AI platform in the world, and you start to understand just how massive this crisis really is.

Not the billionaires building these facilities.

Water costs are expected to rise dramatically across affected regions in 2026 as AI consumption spikes. Businesses and consumers nationwide should expect to pay significantly more for water as data centers compete for the same limited resources. 

Regular Americans — farmers, families, small businesses — are already feeling it. The Georgia story is just one example of what’s quietly happening in communities nationwide while the tech giants write the checks and the politicians look the other way.