Award-Winning Journalist & Speaker - Expert in ERISA Fiduciary, Child IRA, and Hamburger History
Did you ever have a dream you kept putting off? A place you always wanted to visit? A story you always wanted to tell?
So did I. (Notice the past tense.)
This site might give you a clue about how I accomplished this. Who knows? It may even reveal to you how you can realize your own greatest goals.
Interested in learning more? Find me on Twitter and LinkedIn. You can also subscribe to the RSS feed.
Copyright © 2026 Pandamensional Solutions, Inc.
You cannot copy content of this page
Which Town Will Win The Data Center Jackpot?
First, let’s review what all the buzz about data centers is about.
It turns out, more folks have learned how to use Generative AI (GenAI) faster than expected. You may be witnessing the fruits of these labors in short-form videos. Word is, someone is making an entire movie using AI-generated video. This isn’t traditional CGI, or computer-enhanced animation. This is entering a prompt, pressing a button, and—Bingo!—out comes a polished scene ready for viewing.
OK, it’s not that easy. Practitioners explain they may need to tweak the prompt several times before the scene comes out the way they imagine it should. And, by “several times,” they might mean sixty times or more.
That’s a lot of compute. And where does all that power come from? Data centers.
More pervasive and less noticeable, however, are the behind-the-scenes tasks companies of all shapes and sizes are accomplishing with popular GenAI platforms like Grok, ChatGPT, Claude, and Gemini.
I’m not kidding. Almost anyone can do this. For example, I’ve always dreamed of creating an app that helps people determine what their true favorite Beatles song is. It uses an old market research trick called a paired comparison test. The app presents two songs, and you pick which one you like best. After a hundred or so paired comparisons, the app can estimate which song is your favorite.
Spoiler alert: Since the Beatles catalog contains hundreds of songs, you may need to try more than a thousand paired comparisons before you can be reasonably confident you’ve found your true favorite.
After reading that Claude is an excellent choice for “vibe-coding” (i.e., describing what you want rather than writing the programming code yourself), I decided to jump into the deep end. In two or three sentences, I instructed Claude to write the code for the Beatles app.
It worked. First try. I was amazed. It even threw in the album covers for each song.
Now, don’t get me wrong. As with any new thing, there are a lot of snake oil salesmen out there. I just attended a virtual conference billed as an “AI Marketing Expo.” Over 1,500 people attended the event. The organizers promised three days of “hands-on examples,” but it consisted of 75% sales pitch, 20% fluff, and only 5% hands-on content.
It was more than disappointing; it was a cliché. Oh, and did I tell you? The product they were selling costs $10,000. A year.
You don’t need to spend ten grand a year to learn how GenAI can help you market your business better. You can just use ChatGPT, Grok, Claude, Gemini, or any other platform.
And many people are.
The Federal Reserve of St. Louis reported in October that roughly six in ten working-age adults have used GenAI. Exact numbers vary by survey, but one interesting tidbit is that individuals may be adopting GenAI faster than the businesses employing them. What’s more, GenAI adoption appears to be moving three times faster than the adoption of personal computers.
At least as far as people are reporting. The truth is GenAI is becoming pervasive. You may be using it without knowing it. Your favorite app (like my Beatles app) may have been created or tweaked using GenAI tools. And there’s still room for more.
Or is there?
Clearly, at some point, the industry will hit a capacity constraint. In some places, it already has. That’s why there’s a need to build more data centers.
Aye, there’s the rub. People want to use GenAI. They just don’t want to see data centers in their backyard. It was the same issue we had three decades ago, with the advent of reliable, cheap cell phones. Everyone wanted the convenience of the phones. They just didn’t want to see cell phone towers in their backyard.
We overcame the cell phone tower controversy. We’ll overcome the data center controversy.
Probably in the same way. Companies paid landowners when they constructed cell phone towers on the owners’ property. The controversy arose when the neighbors found out. They didn’t get paid, but they had to look at the tower. It’s the same with the solar farm controversy. Landowners get the money; neighbors get the eyesore.
This could be the clue to solving the data center controversy.
Steve Schultz told the Sentinel that, if you remove tax incentives, a full-scale data center could generate enough tax revenue to reduce property taxes in Henrietta by 20%. And that includes more than just town taxes. School and fire district taxes could fall, too. Essentially, all the neighbors (and beyond) would get paid.
Granted, Henrietta doesn’t have room for one of these massive hyperscale data centers. But Schultz estimates that a smaller facility assessed at $250 million could still reduce property taxes by 5%. There’s one catch. The developer can’t receive substantial property tax breaks. Incidentally, Albany’s state-enforced moratorium applies to certain hyperscale data centers, not necessarily to smaller ones that could be built in Henrietta.
This was too good an opportunity to pass up. Henrietta adopted new regulations to accommodate data centers while restricting their impact on local utilities. The town encourages cooling methods that don’t consume large quantities of water. It also requires data centers to supply their own electricity or expand the grid sufficiently to offset their consumption. In other words, existing residents shouldn’t have to pay for the additional infrastructure.
Except for maybe one part. Henrietta prohibits fossil-fuel generators except for emergency backup. That’s too bad.
There are four factors (maybe five) that make a location ideal for building data centers.
Western New York (that’s us!) has all of them.
It has an established electrical grid and high-capacity transmission infrastructure. It has a cool climate and access to abundant freshwater for cooling. (Whoever thought it was a good idea to build water-intensive data centers in the Arizona desert has to be crazy—better to put solar farms there than data centers.) It has access to natural gas infrastructure that could support on-site power generation. (This is the one issue with Henrietta’s code.) It has an established electrical grid and high-capacity transmission infrastructure. And the area contains industrial brownfields that developers could reclaim and return to the tax rolls. (This one is ideal, but not required.)
The best news of all? You can find many of these advantages along the Thruway corridor that runs through several of our towns.
That only leaves one question.
Which town will be the first to win the data center jackpot?
Related