I'm recording now, so you just want to state your name and then your position at clean Wisconsin.
Sure, Amy Barrio, Communications Director with Clean Wisconsin.
Cool, so we've seen Wisconsin be a hub for a lot of data centers and companies like Microsoft
to come here and build these data centers. So Microsoft is building in Mount Pleasant,
making plans for Kenosha, there's Foxconn in the past, and even this week I read that Meta is
actually making plans to invest nearly a billion dollars in Central Wisconsin.
So why are these companies looking at Wisconsin? What makes it so special?
Well, I think when we think about Wisconsin, we have to think about water, right? We are on
two Great Lakes. The Great Lakes is the largest freshwater system in the world,
and these companies may be maybe being a little bit quieter about climate change and
meeting their climate change goals and things like that than they used to be,
but when they're making business decisions, they are very much thinking about climate change.
And so when you look at Wisconsin, compared to some other states, we are projected to be
relatively water-rich in the coming decades, and data centers, particularly AI data centers,
need a lot of water. So we're seeing these companies coming into Wisconsin and other states in the
Midwest, and the draw is our water resources. Those are essential for AI data centers to operate.
Sure. So this might not be something you know off the top of your head, but how much energy
I'm sorry, before you start, could you just brush the top of your head on that side?
Just, I can see a couple here standing up, and I don't want you to-
I don't know what you're saying to me later.
Not sure which side we're talking about.
You're right hand.
Yeah. And one more?
Perfect. I think it's good now.
Okay. Thank you.
Okay.
A little makeup.
Wow.
It wasn't-
You'd read nobody would ever tell me.
They would be like, whatever, tune a TV interview.
Yeah, so how much energy would a data center or location like Mount Pleasant consume on an annual basis?
How much energy?
Yeah.
I don't have that written down.
There's no problem.
But like you mentioned, along with these data centers are going to require water
to cool them down.
So what specific water resources are they going to be pulling from?
So when data centers go through all that computation, they get hot.
And so water is used for cooling, basically evaporation to cool the data center.
And that's typically potable water.
So it has to be clean water at the same level that you would drink.
So oftentimes these data centers will pull from the drinking water system,
or they will drill their own well to access the drinking water aquifers that are below our feet
in Wisconsin.
So these are critical resources.
When I talk about the aquifer, you may not realize we have a big sandstone aquifer
below our feet across much of Wisconsin where a lot of communities draw their drinking water from.
So that aquifer takes a while to recharge.
So in times of drought or water stress, the aquifer levels drop.
And we need to pay attention to how much withdrawals are happening.
We have a lot of competing interests for that water, communities, farming, businesses,
manufacturing, and now AI data centers coming in to access some of that water.
Usually it's water coming from the drinking water system that everybody relies on.
And why does that water have to be potable or fresh water?
Why not salt water or other kinds of water?
I'm not exactly sure.
I think it just has something to do with the components and keeping everything clean and functioning.
Sure, absolutely.
So what are some of the environmental impacts of these data centers continuing to pop up
around was constant pulling this water?
Well, we're trying to see what those environmental impacts are going to be.
I know that the Microsoft Data Center, for example, has said that they're going to use
water recycling reclamation so that they can try to control their water footprint.
We don't have as much information about what these other data centers are going to be using.
And then we share these resources with other states.
So we know a lot of data centers are coming to Michigan, for example,
AI data centers.
So that means big energy, big water use.
And we have to share our Great Lakes water resources.
And what we're not seeing is a lot of communication between states about what projects are being
greenlit, what the water withdrawals are going to be, and what the potential stressors are going to be.
So I think we're in a position where we don't know what the environmental impact is going to be
until it starts to happen.
And that's not a situation we want to be in.
Yeah, thinking of that, the Great Lakes Compact started in 2008, between eight states being able
to manage water diversions.
So can this kind of compact or protection prevent companies from coming in and using more water?
Can this be able to protect the water?
I think it could.
I don't know that it's been used in that way up to this point.
So I think it's just a matter of states kind of coming together and figuring out, okay,
what's happening?
And what you're seeing in Wisconsin in other places is that even local communities'
neighbors don't know that a data center is coming until it's already been greenlit.
And so when we hear about these data centers, it's not, hey, what do we think about this?
It's, hey, this is happening.
And so that's what we're seeing going on in Wisconsin, in Michigan, in a lot of other states.
And it's one of those things where, you know, we have a responsibility to protect the largest
freshwater system in the world.
And we don't know what the future holds with data centers.
I think everyone feels like one second you woke up and AI was here.
And a lot of people don't really realize the impact that this kind of heavy computing is having
on our environment and could potentially have on our really critical water systems.
I think people, you know, there's a lot of excitement around AI, right?
A lot of excitement around data centers coming, you know, what could mean jobs and revenue coming
in. But we have to understand that we have a lot of entities that depend on water resources,
our agricultural industry here in Wisconsin, our tourism, our fishing,
many manufacturing and businesses rely on water and, of course, all of us, all of our communities.
So this is something we have to share.
And so even though there's this excitement to just, you know, go ahead and greenlight project
after project, we can't put those other things at risk.
Absolutely. Talking to someone she had mentioned that in a South American country,
believe it was Uruguay, that a situation come in where a company had built all these data centers
and when you use a couple of this water during a drought year, a lot of the people didn't have
access to that fresh water. Seems, you know, something down the road from something that could
happen in Wisconsin, but is that something people are thinking about?
I think it is something people are thinking about. So that was a Google data center.
And one thing, even when I talk about Microsoft and their effort to recycle water,
they have said that the days when they all use the most water are the hottest, driest times of
the year. Well, that's when everybody uses the most water. That's when our farmers need water.
That's when our communities need water. So when data centers need the most water,
it's the same time when everybody needs most water and that's when you start seeing
the stress and that's when you start seeing people get angry. If you're in a drought,
if your water resources are stressed out and you start having withdrawals for data centers,
it starts to strike at, you know, the real essence of what people understand that they need.
And we cannot say that Wisconsin isn't going to experience those things. In 2012,
we had a devastating drought in Wisconsin. Our aquifer levels were very low. Our water systems
were stressed out. If we have another one of those times, we need to understand that public
needs to understand what AI data centers are going to mean for that. Will they be able to,
you know, slow things down to take some things offline to give our water systems a break? It
doesn't appear so. That doesn't seem to be the business model for AI and for data centers. So
what we really need is an understanding of what could happen if we're in a stressful situation
like that. And we know globally we're at a time when our water systems are stressed and
Wisconsin is not going to be exempted from that forever.
Yeah. Speaking of that as well, Microsoft has plans that says to be water positive by
2030, producing, giving back more water than it is taking in 2022. Their usage globally was,
you know, over 1.5, nearly 1.7 billion gallons of water. That was a 34% jump from the year prior.
So how realistic are these water positive promises? Are these trustworthy?
I don't know exactly how that plan would work to be water positive. It's good that companies are
thinking about this and trying to talk about it and be more public because it's not really a
matter of trust. It's a matter of transparency. So we need to be able to see what are the plans?
How much is this data center going to be using next year or in five years? What are the plans to
make this better? What we're seeing though isn't that transparency always. We're seeing data centers
approved, construction underway, and then we're hearing maybe how much water is going to be used.
And I think the public understands how critical water resources are. They understand that we all
deserve and need safe clean water every day all the time. And there needs to be some oversight
of that, some real understanding of what AI means for those systems. I think it's good that
Microsoft is looking to try to solve this issue because it is an issue. What we're seeing though,
are data centers popping up and approved and getting built before we understand how these
issues could possibly be solved. So we're seeing the issue happen and then trying to figure out
how do we solve the problems that it's bringing. And that's backwards. And we still have a tendency
to do a lot of things that way in our society. We put a bunch of chemicals out there and then
we realize, oh, those were bad. How do we fix that? We put things out and then we have to go back
and try to unwind everything. We know we don't have time to do that with our critical water
resources. We cannot afford to stress things out even more than they're already being stressed
under climate change and all kinds of other issues. Yeah. So do lawmakers have a responsibility
to make regulations and keep track of the water and then how it's being used for these data centers?
I think the public expects our government and our lawmakers to take care of these water
resources. We expect clean, safe water because we know that everybody relies on it. There's not
one person in this country, not one business, not one anything that doesn't need to have water.
And so yes, there's an expectation, I think, that every level of government takes some
responsibility for protecting those water resources. And it doesn't mean that we can't have data
centers that we can't have AI. It just means that we kind of need to take a minute and find some
real understanding of what an AI data center is going to mean for water, for energy use, for our
climate goals. What are these, this kind of new era, what's it going to mean 10 years from now,
15 years from now? Are we going to look back in 25 years ago? Oh my gosh, look what happened
and not being able to do anything about it. Yeah, when we talked before on the phone,
you mentioned a little bit that there could be impacts on taking water from other resources
like farming. How are taking water resources going to, how could it directly impact Wisconsinites,
some of the water usage that they care about? I think, you know, water systems are all interrelated.
So when we think about, for example, I believe in 2024, agriculture had the highest amount of
water withdrawals from our aquifers since they started recording that. So we know that our water
resources are something we have to pay attention to. When you get into a drought situation,
a hot, dry period, it's harder and harder, for example, to reach water. If it's,
aquifers levels are low, it's more energy intensive. It's just puts a lot of stress on everybody,
everything, if you have a private well, trying to get that water. So I think
when we talk about data centers, I can't remember what your question was.
No, no problem. You can ask it again. Yeah, yeah. How could it impact every day
Wisconsinites if, you know, they're not having access to clean fresh water, putable water?
I think everybody knows how special having clean water is. We cannot take it for granted.
When you see communities, even here in Wisconsin, that don't have clean, safe water, for whatever
reason, maybe they have contamination in their private well. Maybe they're having issues with
their private well, kind of running low because of use in other areas. That's when you start to
see how hard it is to live without that resource when you're having to get bottled water in every
day to drink, when you're having to worry about where your water is coming from, when you're a
water utility that's having to use more and more energy to pull the water up out of the aquifer.
You know, people get uneasy when they start realizing that this resource that means so much could
be harder and harder to get. And we don't want to get in that situation.
So with our water resources, you know, in Wisconsin, there are going to be times when we have plenty
and there are going to be times when we have less depending on where we are in this kind of
uncertain climate future. But communities need to kind of have a chance to at least understand
what an AI data center is going to mean in that mix of different kinds of businesses and people
and farmers and entities that want to also have access to water. Absolutely. That covers a lot
of the questions that I have. Is there anything else that you wanted to talk about that we don't
think we covered? Well, I just I want to stress, you know, that it's it's not that AI or data
centers are, you know, this terrible thing that's coming. It's it's more that we need just time to
understand what it all means and how to make sure we are prepared and that we have water resources
that can handle this kind of big shift that we're seeing right now. And the hope is that yes,
data centers can be more energy efficient, more water efficient sometime in the future,
but that's not what we're seeing right now necessarily. So it's understanding,
supporting that push from tech companies to be more water efficient and more energy efficient
and really just giving communities and people who live in the state some transparency and a chance
to understand when we think about our water resources what the impact is going to be.
Absolutely. Thank you. Okay.