Why AI Data Centers Target Small Towns - Old Fashioned Interview with Prescott Balch

[00:00:00] Speaker 1: for people who aren't aware of you, I'll just say initially that you've been just instrumental in sort of evening the playing field for all these data centers coming into Wisconsin. So much of like what people know is straight, ripped straight from the politicians mouths, which is ripped straight from the press releases from the companies that are making billions of dollars. And you've been really great about like, just giving people the full picture. And so that's why I'm hoping that we can do here today. But first I'm gonna make you an old fashioned. So do you like your old fashioned sweet, sour or press? [00:00:42] Prescott: - I'll go with sweet. [00:00:44] Speaker 1: - You're a sweet guy. - Yeah. - Okay. Are you, have you drank a lot of old fashions? [00:00:48] Prescott: - I have never had an old fashioned. [00:00:51] Speaker 1: - He's never had an old fashioned ladies and gentlemen. Prescott, how is that possible? [00:00:56] Prescott: - I gotta, I gotta stay sharp for the fight against data centers. [00:00:59] Speaker 1: - Ah, so you're not a drinker. [00:01:01] Prescott: - Well, I've just never had an old fashioned. - Never had an old fashioned. - I'm embarrassed to say, but. [00:01:06] Speaker 1: - You, where were you born and raised? [00:01:08] Prescott: - South Florida. [00:01:09] Speaker 1: - South Florida. Okay. And what brought you up to Wisconsin? [00:01:11] Prescott: - A job. [00:01:12] Speaker 1: - What job? [00:01:13] Prescott: - I was a technology executive at US bank. [00:01:16] Speaker 1: - Wow. Tech executive at US bank. And that's actually, I'll shut up and start making your old fashioned now. But that's part of what really got you into this, into this fight. It's just, you're, you're knowing what's actually going on at these various tech companies, right? [00:01:34] Prescott: - Absolutely. I knew a little bit from my career. I spent a lot of time. I was on the software side of the business, but obviously all of our software runs in data centers. So I know a thing or two about those data centers. And I also know a thing or two about fortune 50 companies. [00:01:49] Speaker 1: - What's a thing or two you know about fortune 50 company? Now, okay, for people who don't know, 'cause, and when I say for people who don't know, I mean, I don't even know. Fortune 500, that's like the top 500 companies. So fortune 50, those are the top. [00:02:02] Prescott: - The top 50, the biggest of the big. [00:02:04] Speaker 1: - That makes sense. That was too much bitters. Don't worry. That one will be mine if I remember. [00:02:08] Prescott: - I'll never know. [00:02:09] Speaker 1: - Yeah. We always put bitters on a, on a sugar cube to start and then muddle it. So muddling in South Florida, you know, I don't know. I was going to say something about like the Everglades or an alligator, but I lost the analogy halfway through, but it's a thought that counts. You know what I mean, Prescott? - That's right. - So fortune 50 company then. [00:02:29] Prescott: - Really big, you know, and they're well funded and they have lots of lawyers and they have lots of marketing people and they're really good at telling stories. And they're really good at telling their side of the story. - Yeah. - And my goal throughout all of the work that I've been doing across the state has just been to try and trying to fill in the blanks and tell the other parts of the story so that communities know what they're getting into when they are looking at these proposals. [00:02:55] Speaker 1: - Well, and that's, that's what I can't thank you enough for doing because the whole goal here, my whole goal with all of this data center stuff is for people to be informed and then make whatever decision they want to make. - Exactly. - And you know, when I put out the first video, my goal was just put a spotlight on all these people being ignored in Port Washington, 'cause it seemed like they were really getting screwed, but I've never seen so many people say they didn't want something, and then have the common council turn around and say, unanimous support. It just, it seemed like the farthest thing from like democracy or whatever we're pretending to do here. You know what I mean? And so it's just great that you've been able to come out and how did you get involved in this whole data center thing in Wisconsin then? [00:03:47] Prescott: - Well, we had one show up in my backyard in Caledonia. [00:03:50] Speaker 1: - Oh yeah, Caledonia, beautiful this time of year. [00:03:53] Prescott: - It's beautiful every time of year. - Yeah. - We guard our open space vigorously. [00:03:59] Speaker 1: - Shoot, sorry about that. [00:04:01] Prescott: - That's okay. - Yeah. [00:04:02] Speaker 1: I also have ADD and I'm not a good bartender. So this drink may take a while, but I'm trying to, I'm listening to you first. [00:04:08] Prescott: - I can wait, it's fine. [00:04:08] Speaker 1: - Okay, all right. [00:04:09] Prescott: - So we had Microsoft show up with a proposal in Caledonia and they wanted to do it across from the Oak Creek power plant. And fortunately, some people in our community saw the signs early and did a little research, figured out what it was about. And I got involved with a community group and said, you know, I know a thing or two about technology and big companies and I'd like to help. And I took a look at what we were trying to do and did what I'm doing with every other community. And that is just try to tell the story about what the risks are associated with such a big project in such a small town. [00:04:44] Speaker 1: - So for people who are just catching up with us, not really sure what we're talking about. What is an AI data center, a hyperscale data center, and how does it technologically differ from your standard data center for emails, credit cards, all that sort of stuff? [00:05:02] Prescott: - Well, first of all, the term hyperscale just means really big, nothing more than that. It's just, it's massive, anywhere from 300,000 to a million square feet of space. [00:05:12] Speaker 1: - That's huge. [00:05:13] Prescott: - It's massive. [00:05:14] Speaker 1: - That's like, I mean, sorry, I keep cutting you off. [00:05:17] Prescott: - That's okay. [00:05:18] Speaker 1: - But I mean, that's like- [00:05:19] Prescott: - I have a metric for you. - Okay, yes, good. - We have a Kmart building in our town that has been abandoned for the last several years. The one data center was 42 times the size of the Kmart building. [00:05:35] Speaker 1: - Of a Kmart? [00:05:36] Prescott: - Or a Kmart? It's, they're just- - What about TJ Maxx? - Probably more than 42. - 42? - Kmarts were pretty big. [00:05:44] Speaker 1: - How many Lambeau fields is that? [00:05:45] Prescott: - I haven't done that math, but I should. - You should do a Lambeau field. - Yeah, that would be good math to do. [00:05:49] Speaker 1: - Okay, just for the video. The AI data center in Port Washington is- - 27 and a half. - Lambeau fields. - Yeah. - And then I'll just put the number in afterwards. [00:05:59] Prescott: - I'll do that math for you. That's a really good idea. [00:06:01] Speaker 1: - That's cool. [00:06:02] Prescott: - Your question though, what separates an AI data center from a data center is how it's being used. - Thank you for reminding me of my question. - That's fine. I'm trying to keep you on task here. - I appreciate that. I need that. - I'm like your ADHD medicine. [00:06:15] Speaker 1: - Thank you. Yes. Should've taken that today. [00:06:19] Prescott: All right. - When somebody wants to use a data center for artificial intelligence, they're just simply putting different chips in their computers. [00:06:25] Speaker 1: - Oh, I know. GPUs. [00:06:27] Prescott: - That's the right. [00:06:28] Speaker 1: - Graphics. And those use way more energy than your standard ones, right? - Indeed they do. - Which is how we're in this pickle of people paying for these billionaires to become trillionaires. [00:06:38] Prescott: - But they would be using a lot of energy, even if they were regular data centers with regular chips, it would just be less than a ton. - Okay. [00:06:46] Speaker 1: So they're using- [00:06:47] Prescott: - Three quarters of a ton, maybe. [00:06:49] Speaker 1: - I actually looked up the actual unit. It's a ton of power. - That could be. - I looked it up and it's a standard unit of measurement. So they're using a ton of power. - Yes. - And why are they going to, seems like they're going after small communities that can be easily infiltrated. And that might be cynical. [00:07:14] Prescott: - I don't, I don't, I don't think it's cynical. It's what got me involved in this and so engaged with my personal time to try to help. Because I think those small communities are in over their heads on this. And I think the large corporations realize it, whether they did it on purpose or not, I'll leave that aside. But- [00:07:35] Speaker 1: - I'm going to say maybe they did. [00:07:36] Prescott: - I think at the very least they're okay with the information disconnect. [00:07:43] Speaker 1: - Yeah. They kind of benefit from nobody knowing what the hell is going on. [00:07:46] Prescott: - That's right. And not understanding how to negotiate deals and not knowing where the leverage is and just the information disconnect is horrible. And that's really what I've been trying to do is level that so they can make the decisions. But I've now, because I seem to be having contagious ADHD, forgot what your original question was. [00:08:03] Speaker 1: - No, my, I mean, my original question, I obviously remember my original question. - Small towns. [00:08:08] Prescott: Why small towns? [00:08:09] Speaker 1: - Why small towns? And before you answer that, how much, how drunk do you want to get? [00:08:13] Prescott: - Not too terribly drunk. - Not too terribly drunk. - Yeah, I've got to stay sharp. [00:08:17] Speaker 1: - One finger, two fingers. - One would be good. - One finger. Okay, there you go. Just a scotch, just a scotch. - That's good. - I'm only going to do one finger too. And so as you can see, we're both just doing one finger here. That might've been, I don't know what happened. You distracted me. - It's the ice. [00:08:35] Prescott: - Anyways. - So they want to go into small towns because that's where they can find the land. They need 500 acres, a thousand acres of contiguous property to build these campuses on. They don't, they can't find that in Milwaukee. [00:08:47] Speaker 1: - Ah, so they need the land too. [00:08:49] Prescott: - Primarily what they're after is proximity to power. - Yes. - Because the cost of extending power is millions and millions of dollars. So lots of land, lots of power, and that's the magic combination today. And that's largely found easily in small towns. [00:09:06] Speaker 1: - Got it. So essentially they're kind of following these power lines. - Correct. - And just saying like, what's the closest place we can get to a power line with some local politicians that we can control the narrative with? [00:09:19] Prescott: - Correct. - Okay. [00:09:21] Speaker 1: - Easy marks. - Easy marks. So how many places now, actually real quick, I want to ask, are you a garnish guy? [00:09:27] Prescott: - What are my options? [00:09:30] Speaker 1: - Your options are cherries and oranges. [00:09:32] Prescott: - Cherries and oranges. - I want an olive. - Yeah. We could do an olive. [00:09:36] Speaker 1: You want to do an olive? - I'll pass on the olive. [00:09:39] Prescott: - Oh, okay. - But the orange. - The orange sounds nice to you? Yeah. [00:09:43] Speaker 1: - All right. I'll do a little nice, cute. Look at that. See that? [00:09:47] Prescott: - Wow, that's good. [00:09:48] Speaker 1: - Isn't that fun? There you go. Have at it. - Okay. This bar is sweet, pristine, nice, beautiful. We're at Natty Oaks, by the way. Are you a fan? [00:09:58] Prescott: - I'm a fan now, yeah. [00:09:59] Speaker 1: - Yeah. It's a great bar. We've done some bellied ups here. It's a podcast where we call in people. Plays in better to my ADD than trying to break down these complicated topics. [00:10:10] Prescott: - Yeah. This is good, by the way. [00:10:12] Speaker 1: Thank you. - Old fashioned. - Yeah. - Wow. This is exciting. - Yeah. Very exciting. - Nothing like an old fashioned interview, Prescott. [00:10:20] Prescott: I appreciate you coming in for it. [00:10:21] Speaker 1: - My pleasure. - So now that we got that done, I can now focus more. [00:10:24] Prescott: - Okay. [00:10:25] Speaker 1: - So yeah, this interview should get exponentially better the longer we go. [00:10:28] Prescott: - All right. [00:10:29] Speaker 1: - So what were you saying? That's a joke because it's funny. I just said it was going to get better and I said, so what were you saying? So they're after these sort of small towns. Of all the ones going on in Wisconsin, what are the cities we got these in? [00:10:45] Prescott: - Port Washington, Beaver Dam, Mount Pleasant, which is a unique situation. Caledonia, Cassville, Mishikot, where we were, and Greenleaf are the ones that, oh, and Kenosha, the other one that I've been working with. - DeForest as well. - Oh, DeForest, sorry. Yeah. I've done so many now, I'm losing track. - Menominee as well. They have proposals up there. - Yeah, Menominee. I'm going out there later this month. Yeah. They're all over the place and they're all perfectly aligned with the maps of high voltage transmission lines. [00:11:20] Speaker 1: - So basically follow the transmission lines and you'll see places of interest. [00:11:23] Prescott: - Yeah. And then start knocking on doors and asking farmers if they want to sell. - Mm-hmm. - And start putting together 600 acre parcels. [00:11:31] Speaker 1: - And that's sort of what they did in Greenleaf. Now we should mention that Greenleaf, Caledonia, DeForest, those places have all effectively told the data centers, we don't want to be here. - Right. - Microsoft and Caledonia to their credit said we heard that the people don't want to be here. So we are gone. Not to their credit though. They kind of do have a tendency to sort of sneak in there. In fact, there's a video on your website where you show a Microsoft lawyer sort of explicitly stating that most, these are not good for the communities aside for the short term construction jobs. - Right. - But most communities don't want them. It was sort of baffling to see that. And actually let's just play that clip right now so people can see it. [00:12:20] Speaker 3: - And nobody really wants a data center in their backyard. I don't want a data center in my backyard. So you do get a lot of pushback and you need buy-in from the community because you have to get permits to do that work. And you have government officials that they are elected by the community. They certainly want to make their constituents happy. And so you in turn have to make their constituents happy so that you can still get your permit and you can still operate there. Data centers, once they are operational, don't bring a lot of jobs. They do on the construction side, but you're not really getting a ton of that community benefit from having a data center really truly in your backyard. [00:13:00] Speaker 1: - So what was your reaction when you first saw that clip from the Microsoft attorney? [00:13:03] Prescott: - Obviously I knew all along that that was a reality. These aren't job creators post-construction. They're no real benefit to the community. It's always nice to hear someone that shouldn't be saying it out loud saying it out loud. I was actually happy that they said it out loud because it's one of the misleading marketing tactics that the industry is using to try to sell data centers to small communities. And so it's really great when you can take their words and say, you know, here's the truth. These really aren't that good for you. - Right. [00:13:37] Speaker 1: And you know, what's a community that is like, you see a data center moving into this particular community, you're like, not a good idea. [00:13:47] Prescott: - Small. - Small. - And you know, I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. I think that's a good idea. - Yeah. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - That's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - Yeah, I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. - I think that's a good idea. [00:14:37] Speaker 1: - I think that's a good idea. - I think that's a good idea. - On the plus side of an automotive plant, it's at least there for a while and it employs a lot of people. [00:14:44] Prescott: - That's right. [00:14:45] Speaker 1: - The difference here with these hyperscale AI data centers is they employ very few people. - Yes. [00:14:50] Prescott: And I know this will shock you, but the industry lies about that statistic all the time. [00:14:56] Speaker 1: - What statistic in particular? [00:14:58] Prescott: - The employment numbers for a data center. [00:15:02] Speaker 1: - What do they typically say is the number of people? [00:15:05] Prescott: - They'll tell you 200, 400, 600 jobs for a campus and it's more like 20 per building. So we were going to have the Caledonia example. We were going to have three buildings and they were talking about a few hundred jobs. It'd be more like 60. [00:15:22] Speaker 1: - And are these high paying jobs? [00:15:24] Prescott: - Probably half and half. You know, the industry a long time ago figured out how to manage computers remotely. There are very few tasks that require hands-on activity. So you're going to have very few people. And data center industry also has a saying called lights out operation. Their goal is to not have anybody in the building. So your employment is going to be security guards and janitors and a few high tech staff. And that's it. And it will always be that way. And the data center developers who are in front of all these small towns are telling them just straight up lies about how much post construction employment there will be. [00:16:03] Speaker 1: - And so, well, that's very interesting to know. Of the data centers, of the data centers that are currently either in the works or being proposed in Wisconsin, which one would you say has kind of the worst deal? [00:16:20] Prescott: - Oh, Port Washington, easily. That's a horrible deal. I've personally never seen a worse deal. They're giving away $450 million worth of tax breaks over 20 years. They won't see a dime of revenue into the city's general fund for 20 years for a really long, complicated reason. They won't be able to get any economic value out of the deal because of the way state law works. It's just up and down. It's a horrible deal. And I don't know how their elected officials approved it. It makes no sense to me. [00:16:58] Speaker 1: - Well, so what do you mean $450 million of tax breaks? Why would a small city like Port Washington give Oracle, OpenAI, Vantage data center, these trillion, multi-trillion dollar companies, why would this small city of Port Washington give them tax breaks? [00:17:15] Prescott: - Because they want them to build in their communities. And they want them to build in their communities because there's another state law that says if the city wants to raise its tax revenue to pay for more cops and more firefighters and whatever, you have to have construction numbers in order to justify raising the tax levy. So you've got city officials that are chasing that construction at all costs. And the data center developer comes along in the perfect storm and says, I'll build here if you make it worth my while and you give me these tax breaks to pay for all of the stuff that I'm going to do for you. I'll pay for roads and pay for sewer, but you have to pay me back with tax credits. And the city signs up for it because they just want the construction. They want the construction so they can raise taxes on people. Remember when you were in school and your teacher said you should pay attention in math class? - Yeah. - She was right. You should have paid attention in math class. - I know, because if I did pay attention, I wouldn't even need you here right now. [00:18:16] Speaker 1: - I know, exactly, exactly. [00:18:17] Prescott: So maybe it was good that you didn't pay attention in math class. But yeah, if you follow the math and you follow the money here, you can understand why the city is making really bad deals. Because they're incented by state law to do something to drive construction in their cities. And the developers are here because they get huge sales tax breaks. - Which was pretty dumb. [00:18:42] Speaker 1: I mean, I personally, if I were to say in 2023, they said, hey, these trillion dollar AI data. Let's take the richest people in the world and say you don't have to pay taxes on any of the software or hardware. - It's crazy. [00:18:54] Prescott: And you and I, if we go into Best Buy and buy a computer, we have to pay sales tax on it. It's sort of unfair. [00:18:59] Speaker 1: - Take me back actually, why now like Port Washington is every other city. Explain this $450 million and like get into the weeds of that. I don't quite understand that. [00:19:11] Prescott: - Okay, so Vantage, the developer there, says to Port Washington, we will pay up front. We'll borrow the money because Port Washington can't borrow this much money. We'll borrow all the money to do the infrastructure work, water, sewer, whatever that needs to happen there. Roads around this thing. That was a hundred and something, $170 million. Port Washington said, okay, fine. You pay that up front and we'll let you pay yourself back with tax credits over the next 20 years. And they also agreed to pay them back at 7% interest over that 20 years. And all of those tax credits add up to $450 million worth of tax breaks. Now, if they didn't agree to that, if they just said, we'll pay you the 170, no interest, then it would be 170. Or if they had said, hey, Vantage, pay your own way. You need all that infrastructure you pay for, we're not giving you a dime back. Then that number would have been zero. But as I said, the city is eager for this construction because they're trying to solve another problem. And they will agree to anything at any cost. And the developers know it that they can, I think they know it, that they can sort of extort the best possible deal out of them possible. So they got them to agree. Vantage, you pay up front because the city can't borrow that kind of money. We'll pay you back not only the 170 million, but we'll do it at 7% interest over 20 years. And that will over that 20 years will be $450 million worth of tax breaks. [00:20:42] Speaker 1: It's kind of like your second cousin, putting a meth den on your hunting property and then asking you to pay for the beakers. Yeah, pretty much. [00:20:52] Prescott: Okay. [00:20:53] Speaker 1: Yeah, that works. That works. Right. They must know. They're smart people working in the Port Washington government, you must think, right? You didn't get there being dumb. Do they not know that this is happening? [00:21:11] Prescott: It's an interesting question. I've tried to figure that out. They've admitted that they have a problem trying to get value out of their deal. So to me, that means they knew going in the problem that they had and that they were hoping for some legislative change in order to rescue them from the problem that they have. I honestly can't walk a mile in their shoes and figure out their logic. I just don't get it. It's a bad deal. It's going to cost the city a lot and everything has to go perfectly for them to even come out of it in 20 years where they think they are. And nothing goes according to plan over 20 years as project plans fall apart in year one. So I honestly don't know. I wish I could understand it, but I just think it's a terrible deal and I don't understand their logic. [00:22:03] Speaker 1: Are taxpayers in Port Washington going to have to pay more taxes for this data center? [00:22:08] Prescott: Well, in fairness to the city, they have a choice. The choice they have is as the construction from the data center hits the books, Port Washington can raise taxes on people because of the way state law works. But if they do that, only the current residents will have their taxes increased. That would be a really bad message to send to people. So will they do it? I don't know. The other choice that they have, though, is all right, we won't take the construction. We'll leave it and we won't raise your taxes. And then in 20 years when the project is over, everybody's taxes will drop like a rock. Will they do that instead? I don't know. That's the big choice that's in front of Port Washington. And that's what I want the people of Port Washington to understand. You need to watch what your government is doing because they have a very important decision to make about whether to use this project essentially to raise your taxes or leave them alone. And they've been telling you that they wanted to do the project so that they could improve your services. Well, the only way they can improve services is to raise taxes. The only way they can raise taxes is to raise your taxes. And I just think they're not being straight with their community about the challenge that they have in front of them and the tough decision that they have to make. [00:23:25] Speaker 1: And let's take almost this tax issue aside. The idea of paying the interest over time. I mean, is that typical that a city would be on the hook for paying interest for infrastructure, a multi-trillion dollar company needs to make their data center? [00:23:45] Prescott: I don't know if it's common or not. I found two problems with it, I guess. One, just the sheer magnitude of the bill. But the 7% fixed interest rate over 20 years seems like the city gets no upside if interest rates drop. You're stuck paying 7% forever. That Vantage could renegotiate the terms of the loan to a lower interest rate and pocket the difference. So that doesn't seem fair. But the city officials will tell you, well, it's not real money. It's Vantage's tax payments that we're just not taking from them or we're giving back to them. But it is real money because if you weren't doing that deal, that tax revenue would be going into the city funds to pay for those new services that they said they were after. So they talk themselves into, I call them smart people who talk themselves into really stupid things. They've talked themselves into this being okay. That it's not really money from the residents because it sort of never gets collected. It just gets returned immediately to the taxpayer. [00:24:49] Speaker 1: And then sort of branching out from this very local issue to the AI industry more broadly. What is a typical life cycle of an emerging technology? Like we saw the dot-com bubble. Is what we have now similar or different from the dot-com bubble? [00:25:09] Prescott: It's up for debate. My personal opinion is it's sort of exactly the same thing. We go through these cycles all the time in the technology industry. Something new comes along and we start banging the drum about how great it's going to be and how it's going to change everybody's lives and infect every product and so on. We're doing the same thing with AI now that we did with the dot-com era. And in the dot-com era was the internet and that was going to be everywhere. And it is everywhere. It just goes up this big hurdle. We overhyped technology. Do everything. We overbuild. We overbuilt for dot-com. And then everybody's business model was flawed. So all those companies go bankrupt and we go back down into the trough. And then eventually we find steady state. I think the same thing is going to happen with AI, but we're on that way up that crazy line. And we're overbuilding data centers. And at some point somebody is going to say, you know, you make money on this or we're out. And when those funding mechanisms pull back, the industry is going to have a real reset on its hands, just like it did with the dot-com bubble. [00:26:14] Speaker 1: So, I mean, you're from the technology industry. You're from the big business industry. We've all seen these different booms and busts in technology, like the dot-com bubble. There's talk out there of open AI running out of money. Now open AI is one of the people in the Port Washington data center. Like let's say, or in Oracle is another company that is, you know, people have questioned their finances quite a bit. So let's say both those companies like go under or Vantage goes under. Is there any protection for Port Washington if those companies go under? Do they just have a half-built data center? [00:26:50] Prescott: So most of these communities are at least doing the basics about having protection against a big company walking away. Bankruptcy is another issue. And like all good issues like that, the lawyers will get really rich when that happens. And who knows what will happen, it will be up to the bankruptcy courts to figure that mess out. But assuming there's no bankruptcy, there are the basic protections against just simply walking away. The risks really are partial completion of a campus, one instead of five buildings, so they get less revenue. Another risk would be a couple of years down the road, we realize we've built too many of these things and now the technology is more efficient or the people just stopped wanting to fund it because there's no profit in it, whatever. In that case, the village's risk is really a lower valuation on the buildings and less tax revenue. So I'm not going to get a million dollars a year in a tiny little village, I'm going to get a hundred thousand instead. That could be bad if they were banking on a million. There's a whole lot of things that can go wrong along the way that aren't catastrophic, like building doesn't get finished. Technology changes all the time, it gets more efficient all the time, it changes all the time. When a community is looking out 25 years on the timeline and thinking we're going to go race to this dollar amount and then stay there for 25 years, they're not seeing the industry like I see the industry, which is always changing and always trying to get more efficient and always trying to spend less money on stuff. The real risk is just change over that time period and change that will be worse for the community. It will not get better for the community. The industry is doing a great job today telling the communities what the best case scenario looks like and they're not telling them what the downsides are. [00:28:39] Speaker 1: I mean, is there a use for these hyperscale AI data centers if like say quantum computing means that you go from this really big data center to like maybe you don't even need data centers or they're talking about putting data centers in space and everyone decides they're just going to put them in space. Is there something else you can do with these data centers in the tech world? [00:29:05] Prescott: It was one of the arguments we made in our town. What are you going to do with that thing when it's not usable as a data center anymore, economically usable as a data center anymore? And they're very purpose built buildings. They're hard to think of. It's hard to think of a reuse opportunity for them. They're low ceilings. There's no windows, lots of power, and they're massive sizes. What is a small little town going to do with 3 million square feet of hard to reuse commercial real estate space? They're going to look at it and regret their decision is my opinion. [00:29:36] Speaker 1: Or, or they might go, what if we do 600 yard bowling? Could be. [00:29:43] Prescott: Yeah. Or you could, you could go the other direction and have 200 lanes. [00:29:47] Speaker 1: Oh yeah. With bumpers. Probably. [00:29:50] Prescott: Yeah. Yeah. You can sort of have a lane for every family in the town. [00:29:54] Speaker 1: That could be cool. I mean, now I'm starting to come around. Yeah. Right. Honestly, can we just skip the AI thing and just go right to this. Just go right to the bowling alley. That would be cool. [00:30:03] Prescott: Yeah. And the air conditioning will be wicked too. Oh yeah. [00:30:06] Speaker 1: It'd be really. Yeah. It'd be closed loop cooling system. Yeah. There you go. Hey, is that, is that a, is it true that sort of a closed loop cooling system? That seems to be what they're always saying now to kind of tell people this won't use that much water. It's a closed loop cooling system. Is that a, um, like, is that really going to like mean very little water? [00:30:31] Prescott: It's going to mean a lot less water than the previous generation of cooling. [00:30:35] Speaker 1: Mm-hmm. [00:30:36] Prescott: There's some question about what they put into the water and how they dispose of it after they're done with it. Cause it, the water will only last so long before they've put so much stuff in it that they have to physically remove it more importantly for that argument. It's, it's an example of how the industry is always changing. It was not, but three years ago when there was no such thing as closed loop cooling. And now if you're not building a data center with closed loop cooling, you're not building a data center. Cause the community will tell you, you're not using that much water. And now everybody that's got the open loop is getting looked at like you guys need to retrofit to this new thing. Cause we want you to use less water. It's an example of how quickly the industry changes and how delicately balanced these things are from tipping over and becoming no longer viable because the cost of retrofitting the new technology is too high. [00:31:25] Speaker 1: And I think it's also worth saying like, even though it's a closed loop system, those take more energy. That's right. And then all those offsite energy plants, those still use a lot of water. Correct. So there's still a lot of water being used if not just onsite. Yep. And then what about everyone's power bills? [00:31:44] Prescott: Yeah, this is a fun topic. I have my own opinion on it that the power companies do not, this is hard to believe too. They're not very transparent. [00:31:54] Speaker 1: Power companies? Utilities? [00:31:57] Prescott: Yeah. Public utilities? I just learned this. [00:32:00] Speaker 1: For the public good. [00:32:01] Prescott: Right. [00:32:02] Speaker 1: Are they really not transparent? [00:32:04] Prescott: They're not transparent. Come on. So if it were me, king for a day, I'd say, show me your math. Show me how you calculate this tech company's data center's bill. Show me the algorithm that you use. Give me a spreadsheet. I'll go get some experts to criticize it. They don't have a spreadsheet that they show. They don't show their work. So the whole model is built on trust me. And until I see the math, how do I know that they're actually going to charge them actual costs for the electric bill for a data center user? I don't know, because they don't show their work. [00:32:44] Speaker 1: Well, and so in the new bill that was proposed by Wisconsin lawmakers, number one was they said that the rate payers will not pay more money for these data centers. Is the way that law is written actually protecting rate payers? [00:32:59] Prescott: Only if you believe in pinky swearing as a way to prove that that happened. [00:33:04] Speaker 1: Well, I just so happen to believe in pinky swearing. [00:33:06] Prescott: Well, then you're good. [00:33:07] Speaker 1: I do. I trust, actually, no, I pinky swear with my brother and he ended up telling my parents where I was hiding the red dog. So I'm not sure that that's such a great idea. But so really what that bill does is it says, trust me, bro. Yes. [00:33:26] Prescott: Yeah. [00:33:27] Speaker 1: Pretty much. And that's, it says though, it says have the PSC shall be responsible for making sure that no people. [00:33:37] Prescott: PSC will pinky swear with the energy company. Yeah. [00:33:40] Speaker 1: Okay. So there's no way that we can really have transparency coming back again to transparency. Yeah. We don't know how that's happening and how can we be confident that our bills are not being raised due to this. [00:33:53] Prescott: Right. [00:33:54] Speaker 1: Okay. Interesting. Let's take a step back real quick. What is like when most people think of artificial intelligence, they think of ChatGPT or they think now every time you Google something, you've got this like AI thing giving you a sometimes right, sometimes wrong answer. It seems like it's kind of being pushed down our throats more than it is we're demanding it. Is that fair to say or not really? [00:34:21] Prescott: It's fair and it'll, it'll only last as long as that industry can figure out how to get paid for it. So today they're pushing it on you because they think it's helpful. They think that you want it. And as long as you keep using it, they'll keep pushing it on you. But at some point they'll need to make money off of it. This happened with the dot-com boom would think the internet's a great thing. All these companies are firing up, but nobody knew how to make money on it. And eventually that crashed because they couldn't figure out how to make money. But the industry is doing right now with AI is pushing it on you, hoping you'll get enamored by it. And then, then they'll slip you the bill and say, all right, now we want you to start paying for this. And eventually they'll come up with some use of AI that is of some value that people will pay for. But at this point, it's not tomorrow that that's going to happen. But is it helpful? Yeah, it's probably helpful to some people. Like you said, sometimes it's right and sometimes it's wrong. And sometimes it's going to make us dumber because we don't do our homework on what it's actually telling us. But the industry is just now so early in the process of trying to figure out how to make money on it that they're just pushing it on you everywhere in order to see what sticks. [00:35:38] Speaker 1: It's almost like best case scenario we get for these tech companies, best case scenario, we get super intelligence. And if we get super intelligence, that's kind of like the first one to super intelligence controls everything because this thing is so much smarter than all of humanity. But what that means, like, let's say they achieve that. That just means all the power and financial power in the world goes to one company. So in one hand, and I think, or let's say it maybe it's a few hands, a few people achieve super intelligence. Well, now we're just giving more control over like if these robots are doing all the jobs, these companies are getting all that money. They're talking about universal basic income. So it just comes down to now we have to trust these tech companies. But unfortunately, I'm not sure the tech companies have given us any reason to trust them. If you look at XAI putting out like child . If you look at Google, how many lawsuits have they been in where they said they weren't collecting our data and now they're collecting it. Instagram says we're not listening to your conversations. And if I say bubblegum big league chew, bubblegum big league chew, bubblegum big league chew. Now everybody at some point is going to get an ad for bubblegum big league chew. I mean, which is a fantastic gum by the way, not a sponsor, but that's what's up. But I mean, they haven't given us any reason to trust them. Even Meta, there was a lawsuit about they were, they put in the language that their AI chatbots, their friends were to create sensual relationships. I think that was their word, something like that with kids, you know, and they specifically put that in. I mean, these guys have given us no reason to trust them. And now with AI, it's almost like they're asking us to give them all the trust. Well, we haven't asked them to pinky swear yet. [00:37:34] Prescott: Oh yeah. [00:37:35] Speaker 1: We got to ask them to pinky swear. Let's have our Wisconsin lawmakers put another pinky swear alarm with the big tech community. [00:37:41] Prescott: It'll be the sixth item in the bill. [00:37:43] Speaker 1: Yes. Yes. Yeah. What a small bill too. Like to make cheese, you got to like go through like a Bible, you know? Yeah. But for, for the biggest industry in the world, two pages, you know, and they, and that's double spaced. [00:37:59] Prescott: I know. Two. The margins are really wide too on those. Wide margins. [00:38:03] Speaker 1: Yeah. They, they, they, they got to some tricks to get to two pages on that. Yeah. And one of the, the first point restates the current law. Right. So, I mean, I don't think that should count. I think it could fit on one page. [00:38:14] Prescott: It's like the essay that had to be three or four pages long in high school. And you know, you used all those same tricks to get to the word count. Yeah. You use bigger words. [00:38:24] Speaker 1: Yeah. Get the, just the source out, you know? Yeah. You almost wonder if that, why they didn't just say chat GPT, help me write a bill that's three pages. Right. What are a couple other places? Like let's take Beaver Dam, for example. Um, Beaver Dam, it seems like they had an NDA in place, a non-disclosure agreement. And the people living there didn't even know what was happening until ground was breaking. [00:38:49] Prescott: Yeah. [00:38:50] Speaker 1: Um, are you, how many of those have you come across in your work? [00:38:54] Prescott: I've seen, I've seen way too many instances of non-disclosures. And if anything comes of this, I really hope it's that the local governments realize that's just bad out of the gate. That's bad. Don't do that. These tech companies should find some other way to hide their intellectual property if they're worried about it and need an NDA. But the, the lack of transparency, whether it's with an NDA or whether it's just deciding not to say things to your community that just needs to stop because these, these projects are so big. People, people don't realize how big they are. It's going to double the size of Port Washington. It's going to be a third the size of Beaver Dam or, or close to half the size. It's just massive footprint adding to a community. It's landscape changing. It's, it changes the tax base entirely. And to do that quietly is just a huge disservice to, to your community. And if you got elected to serve your people, not tech companies. So if you're going to change their community, you'd think you'd sort of want to take it slow at the beginning and say, Hey, we're thinking about doing this. How do you feel about it? And, and I frankly don't see that anywhere in any of the communities where proposals have gotten any amount of traction. They are just going for the finish line. Like it was somebody wanting a zoning change for a garage. You know, they're just racing to the finish line as fast as they can and hoping nobody notices. And unfortunately in Beaver Dam, people didn't notice in time to push back. [00:40:25] Speaker 1: Yeah. And you even take some things like in Port Washington, where they had hours. They agreed to hours that they could do construction, which I think was like 7 AM to 5 PM or something like that. And now all of a sudden it's a 24 hour construction cycle. Like it seems like a bit of a bait and switch. I mean, is that fair to say from, from your standpoint, how do you change those agreements so quick? [00:40:54] Prescott: Yeah. Well, it wasn't a bait and switch to me. We argued with our own community and we've said to others, when you sign yourself up for a single taxpayer, that's this big, they're going to own you. And once they move from trying to attract you to being married to you, they're going to change. They're going to demand whatever they want. And you're going to have to say yes. And it's exactly what happened in Port Washington. The very first thing that that technology company did or that developer did after approval was say, we're running behind schedule, which by the way, hurts the city because you guys are reliant on this 20 year plan. We want construction to happen seven by 24. And the seven days a week, 24 hours a day. Yeah. Yeah. And the planning commission meeting where it was discussed was embarrassingly bad. There was no pushback. There was one guy that said, please turn off the lights that you're not going to use when you're not using them. Please try to minimize the impact on residents. Instead of saying, no, you can't do that to the people. Like we're here to serve the people, not the tech company. How on earth would you approve a seven day a week, 24 hour a day schedule if you're trying to work for the people rather than the tech company building a big building on your land? [00:42:20] Speaker 1: Yeah. [00:42:21] Speaker ?: That's it. [00:42:21] Speaker 1: That seems kind of messed up because even, even if you take like what, how it's supposed to work, how the people agree or what the common council agreed to in Port Washington was not a 24 hour construction day, but now they just agreed to it. Doesn't there have to be any public input on that? Apparently not. [00:42:40] Prescott: They just approved it at the planning commission and carried on. [00:42:43] Speaker 1: But is that like legal? Is that? [00:42:46] Prescott: I don't. It's probably specific to the zoning regulations of Port Washington and their ordinances. So I'm not, I'm not sure. Got it. Yeah. [00:42:55] Speaker 1: Yeah. But it does seem messed up that you, as a common council, tell people one thing and then not a few weeks after the ink is dry, you completely change a very critical thing, especially for homeowners around the area who now have to deal with 24 hour construction. [00:43:15] Prescott: Yeah. I thought it was terrible as well, but I can see why they did it. They've got this plan that last 20 years, everything has to go according to plan in order for the city to get anything out of it at the back end. Well, if you fall behind in year one, you're going to wreck that plan. So the developer owns you. You, you bet your whole city on this one project. And if they're falling behind the city council or planning commission or whomever made the decision is going to think, well, we have no choice, but to do this because we bet the farm on this one project. [00:43:48] Speaker 1: How did you get involved with bringing more transparency to this AI data center deal? [00:43:54] Prescott: Well, after we fought back on our proposal in our backyard in Caledonia, I both reached out to other communities that had proposals and was reached out to by those communities. And I just got involved because I dislike unfairness. I dislike small communities getting taken advantage of. So I couldn't, I couldn't stop. I had to help. I had to offer my expertise to help level that playing field of information so they could make good decisions. And I've, I've traveled around the state. I've talked to dozens of communities in multiple states and I'm, I'm just trying to level the playing field. And it's just because I'm wired to dislike this, this taking advantage of small communities. Yeah. [00:44:38] Speaker 1: And you've done a great job. I mean, we did that, that event in Michigan. Um, and really it seems like when you bring communities together and you present them the other side, or frankly, I just think the, the truth of what's happening. Communities have more questions than they have answers. And they typically decide in the case of Greenleaf. And I feel like the Michigan is kind of going this way. Caledonia when presented with the raw numbers and what this means, most communities say, no, absolutely not. What is the playbook for a community where maybe they're just hearing about a proposal for a data center? What, what can a community do? [00:45:18] Prescott: Well, the very first thing they can do is try to get educated. There's a, there's a group in the state, an informal group that is formed from all the different communities. We're, we're relatively easy to find. Um, we're happy to help educate. There are people that have expertise in zoning, people like myself, expertise in the technology and financial risks, people, experts in, in energy. We can help you understand what you should be doing, um, whether it's attending to your zoning regulations or whether if you have an active proposal, just sort of understanding the mechanics and, and the pros and cons for your community. We're ready to help. Um, we think we've got all of them covered, but if there are any others out there, you know, we will either reach out to you or you can reach out to us at any time. We're happy to help. But education is the key. You gotta, you gotta know, you have to, you have to move from, I don't know what I don't know to I know what I don't know. So you can ask the right questions and, and, uh, understand both the risks and the rewards. The tech companies are going to tell you all the rewards and they're going to give you eye watering numbers to stare at. And you have to understand the risks that come along with it. [00:46:26] Speaker 1: And it also seems like, uh, when we did the Michikat, um, event, um, we did an informational meeting in Michikat, we brought together, not just the people of Michikat, but also invited the, uh, local, uh, lawmakers from the surrounding community. Um, uh, the village boards and the common councils, and they did show up and they themselves were like getting this information. So, um, I think also a big thing is like to not, you know, come after your local representatives because a lot of times they know the same that you do. And they're trying to figure out this crazy fast moving technology and almost to treat them not like they're on the opposite team right out the gate. Yeah, absolutely. [00:47:13] Prescott: It was heartwarming to see all of them show up at Michikat. And I have high hopes that they'll learn what they need to learn and they'll make the right decision for themselves. Um, the difficult spots are the places where the city council village boards are actively resisting learning because they just looked at that number and they can't hear anything else. Those are the, those are the troublesome spots. Those are the people that are hard to convince, but you have to try. You have to keep being polite and professional and make the arguments and just help teach them what they're missing on the risk side. [00:47:50] Speaker 1: The other question is, um, what is something that bothers me about these data centers is like, what's the data in the data center? You know, what, what is actually going in? What, how much stuff needs to go in these massive buildings? Where do you get it from? Is it copywritten material that they've taken? Is it our data that a lot of us unwilling, unknowingly are giving them? Either because we signed a terms and conditions that nobody read or because they, in the case of Google data from their, you know, assistant, and it's just always on, always listening. Like, do we know if the data in these data centers is legally taken? We don't. [00:48:29] Prescott: That's going to be the responsibility of the company that's running the data center. If it's an artificial intelligence data center, you can assume it's every piece of data they can get their hands on. Because that's how they build these artificial intelligence models. They just gather as much as they can and then run these really long running programs against them so that they can answer your question about whatever. [00:48:52] Speaker 1: So any piece, they scrape the internet for anything? Oh, sure. [00:48:57] Prescott: Yeah. And they'll, they'll get data any way they can. Because the more data, the better the output. These AI models are reliant on large, large quantities of data. There are other types of data centers that are just sort of rent a data center. Like the one that was proposed in DeForest was a company that was just in the data center rental business. In those cases, the owner of the data center often doesn't even know what's in there. They're just renting space to somebody who's bringing data in. It could be data that's different tomorrow versus the next week. And so you never, you never know what's there. [00:49:32] Speaker 1: And is it, is the, like we see these flock cameras popping up all around and flock cameras that are kind of like surveillance systems for various. There are a bunch in Northern Wisconsin, up North, they got these flock cameras and they're taking in data all the time on cars and identifying people even driving the cars. I mean, this is data that most people don't even know is being taken. And then, so it's almost like we are the product and that they're taking our data. And now they're asking us to lose our communities to store this data that we don't even want collected. I mean, is there anything that people can do if they don't want their data collected in public spaces to just constantly have to give that data? [00:50:22] Prescott: You know, there's unfortunately, probably not an easy way to opt out of all of it. If you're in a community with one of those cameras, I would encourage you to call your community, your village or city hall and say, like, what are the terms and conditions on the use of that data? Are you sharing it with anybody? But even if they give you a favorable answer, it could be hacked. You know, somebody could steal it and use it even though you hadn't consented to doing so. Yeah, it's an unfortunate problem. There's no good answer to it because there's just plain and simply so much data out there. And it's impossible to catalog it all and control it all. And it's just that that can be scary if you think about it. [00:51:03] Speaker 1: Could it be possible with AI to just have a like tool AI, which I've read about, which is basically like we have medical issues and to create an AI system specifically geared toward fixing those medical issues as opposed to just agentic AI, which is essentially like these agents that kind of do everything and sort of take away human jobs and take away people's purpose in a lot of way. Is there a way to get the benefits of AI without getting the social unrest and turmoil that comes with? [00:51:39] Prescott: Yeah, probably. The end user has to make that decision. If you're using tool AI for very specific purposes, very helpful things, and you're using that, that's a signal to the companies that are providing that that they should keep providing it. If you stop using the rest of it, that's a signal to them to stop doing that other stuff. It is, however, no matter what, it's going to be hard to control all of the data that's being captured. There's just too much. There's no, there's no way to do it. There's no way to control it. It's, we'll have to keep fighting, no doubt, a never ending battle, but it's just a gargantuan task to try to control it all. Yeah. [00:52:27] Speaker 1: I mean, it, it, it, it seems like at every turn, the people are asked to pay more, you know, the fact that we've given our data and now they want us to pay with higher tax bills and higher electricity bills for these data sets. Pay with our land, pay with our water. We're like, we're like the lobster at red lobster. And now they want us to pay to rent the tank. You know, it just doesn't feel fair. [00:52:52] Prescott: It's not, it is not fair at all. You are, if you're not, if you're not paying for the product, you are the product as the saying goes. It's unfortunately the world that we live in. And it's one that needs constant, uh, diligence and pressure to at least take the edge off of it. Um, but I don't, I don't see us stuffing the technology back in the bottle. I don't think that's too hard. [00:53:15] Speaker 1: Right. So at, at this time, you know, uh, right now it doesn't look like there's a ton of appetite on the federal level to regulate in any way. In the state level, it seems like there's not a ton of appetite to regulate in a meaningful way. And so it just comes down to our local communities. Is that? That's my take. [00:53:34] Prescott: I've, I've advised every community. Don't look for the state to bail you out. Don't look for the federal government to bail you out. This is a local fight. If you want to push these things out of your community because you don't like the risks that come with the rewards, you need to go after the few people that are going to vote on it in your city council. That's, it's as simple as that. But the, the state government created this problem with its various statutes that create a perfect storm. Um, and I don't, I have yet to see enough understanding of the problem that they created that I would think that they could fix it. Let alone it's an election year and a likely contentious one. Yeah. [00:54:16] Speaker 1: Any final thoughts, things I didn't bring up that you thought would be important to kind of bring up? [00:54:21] Prescott: Um, no, I think we covered it. I think we're good. Yeah. [00:54:26] Speaker 1: All right. Uh, Prescott Balch, ladies and gentlemen, thank you for coming by. My pleasure. [00:54:31] Prescott: Thank you. [00:54:32] Speaker 1: And your first old fashioned, you know, you barely drank it. Now you got to chug it. [00:54:36] Prescott: I'm trying, man. I'm trying. [00:54:39] Speaker 1: Thank you for coming by. You're welcome. Old fashioned life, it gets sweet and it gets sour.