Before I get into this week’s issue, one of my peers is running a survey and needs people to fill it out! It only takes 5-7 minutes. See info below:

“I’m running a national survey on technology adoption, virtual care, and AI in PT and OT practice. The goal is to build a real, data-driven benchmark for rehabilitation professionals.

Everyone who participates gets the executive summary before it goes public.

We’re also doing a drawing for participants:

Take the Survey

There's a growing movement to shut down data centers, and I understand the impulse. The environmental concerns are real, and I'll get to them. But a lot of the loudest voices are operating on a fundamental misunderstanding of what a data center even is.

If you sent an email today, you used a data center. Streamed a show, checked your bank balance, backed up a photo to the cloud, asked your smart speaker for the weather, scrolled Instagram, joined a telehealth visit? Data center, every time. Whenever you send an email, stream a show, save a photo to the cloud, or ask a chatbot a question, you're interacting with a data center. This infrastructure has run modern life for over two decades.

So when someone says "shut down the data centers," what they often mean is "shut down the AI." Those are not the same thing. AI is one workload among many that data centers handle. You cannot pull that thread without unraveling email, banking, telehealth, streaming, and basically the entire digital world we now depend on to work, learn, and receive care.

This matters because our advocacy has to be precise to be effective. "Ban data centers" is a demand that collapses under its own weight the moment anyone examines it. "Regulate their water and power use, require transparency, and protect the communities hosting them" is a demand that can actually go somewhere. Precision is not a betrayal of the cause. It's what makes the cause credible.

This week I want to walk through a genuinely excellent 2026 data center report from Programs.com, because it breaks the whole picture into pieces that are easy to understand, and then translate those pieces into what they mean for our communities. I'll keep it plain.

As of July 2026, there are nearly 12,000 data centers across 179 countries, and about 38% of them are in the US. We have almost 8x more than any other country. Demand is projected to nearly triple by 2030. This is not a passing trend. It's a build-out on the scale of a new kind of national infrastructure.

This is the part that surprises people. A typical data center covers about 100,000 square feet, and the largest run to 10 million. But once one is built, it employs fewer than 200 people to operate. So a community gives up a large parcel of land, and often significant tax breaks, for a facility that creates lots of temporary construction jobs and very few permanent ones. That's a real tradeoff worth scrutinizing at the local level. Advocacy #1: Local employment requirements for data centers.

Data centers do generate serious money. States with clusters of them create more than $30 billion in additional economic output annually. But there's a catch: 37 states grant sales tax exemptions to data centers, and one Microsoft data center built in Illinois pulled in more than $38 million in tax exemptions while creating just 20 permanent jobs. So the "economic boon" is real for some and quite thin for others, and residents sometimes subsidize it without seeing much back. Advocacy #2: Data centers don’t deserve tax breaks and should pay their fair share of taxes, preferably directly into the communities they are stealing from.

Two resources matter most here: electricity and water.

On power: by 2028, data centers could consume up to 12% of all US electricity, up from about 4% in 2023. When that much new demand hits a local grid, someone pays for the upgrades. Often, it's residents. In areas surrounding data centers, monthly electricity costs have climbed by as much as 267% over the past five years, and one Carnegie Mellon study estimates data center growth could raise the average US electricity bill by 8% by 2030, and over 25% in the highest-demand markets. That's a household budget issue, which is a public health issue when people have to choose between the electric bill and other necessities. As an example, my own electric bill rose about 19% per kWh over the past two years. Data centers are one driver of that trend, especially here in a top-three market like Chicago, though not the only one—partly due to grid updates, aging infrastructure, fuel costs, and the growing expense of hardening systems against extreme weather (i.e., climate change).

On water: cooling all those servers takes enormous amounts of water. Large data centers can use up to 5 million gallons per day, and US data centers directly consumed about 17 billion gallons of water in 2023. In water-stressed regions, that competes directly with what communities need to drink and grow food.

One local note, because I live here: Chicago is one of the top three data center markets in the country, behind Atlanta and Dallas-Fort Worth. This is not a far-away issue for those of us here. Our key “advantage” compared to the other two? We have access to a very large body of fresh water, which is less of a perk and more of a target.

The build-out is moving toward rural areas with cheap land and available grid connections, and toward new power solutions, including on-site natural gas plants and, increasingly, nuclear. Where and how the next wave gets built is still being decided, which means there's still room for communities and advocates to shape it.

Everything above lands on the people we serve. Higher electricity bills strain the fixed-income older adults many of us work with. Water stress and increased fossil fuel pollution worsen the respiratory and chronic conditions that show up on our caseloads. And the communities bearing the heaviest burden, often lower-income and historically marginalized, are frequently the same populations already facing health inequities we're trained to address.

AND, notice what the honest version of this argument requires. It is not "AI is evil, shut it all down." It's "this infrastructure has real costs, those costs fall unevenly, and we should advocate for the people absorbing them." That's an occupational justice argument grounded in facts, and it's far harder to dismiss than a slogan. Understanding how data centers actually work is what lets us make it.