Jan. 19, 2026

Who Actually Pays for AI's Environmental Cost?

The player is loading ...
Who Actually Pays for AI's Environmental Cost?

Microsoft announced they'll cover the environmental costs of their AI data centers - electricity overages, water usage, community impact.

But here's the tension: AI energy consumption is projected to quadruple by 2030, consuming one in eight kilowatt hours in the U.S. Communities have already blocked billion-dollar data center projects over water and electricity fears. Is this Microsoft accountability, or damage control?

Charlie Harger from "Seattle's Morning News" on KIRO Radio joins us with mor eon why this matters now:

  • Why AI data centers are losing community support and costing billions in cancelled projects
  • What it actually takes to power AI—and why current infrastructure can't handle it
  • How Microsoft's commitment differs from silence from OpenAI, Google, and Chinese AI companies
  • Whether small modular reactors and fusion energy can solve the problem or just delay it
  • Why this is ultimately a West vs. East geopolitical race with environmental consequences
  • What happens when five of the world's most valuable companies all need the same scarce resources

----

GUEST WEBSITE:
www.mynorthwest.com 

----

MORE FROM BROBOTS:
Connect with us on Threads, Twitter, Instagram, Facebook, and Tiktok

Subscribe to BROBOTS on Youtube

Join our community in the BROBOTS Facebook group

Jeremy Grater (00:00)
the environment. Something we all care about, at least most of us, and it's something we all need to survive.

but with the rapid growth of AI and its increasing need for energy and resources like water.

price tag continues to go up.

big tech company is stepping up to cover the bill, or so they say.

But will the rest follow their lead and how will this impact the AI industry going forward?

That's the focus of today's BroBots.

Jeremy Grater (00:24)
And welcome to BroBots. This is the podcast that tries to help you use technology and AI to be a better human.

And we want to get right to this one today. So let me introduce our guest. He is a veteran Seattle journalist. He has been in the media industry for decades.

He's one of the hosts of Seattle's morning news on Kiro, KIRO radio in Seattle. Please welcome an old friend of mine, the lovely and talented Charlie Harger.

Jeremy Grater (00:46)
Charlie, the big story that we're talking about today is Microsoft basically taking responsibility for these huge data centers saying they're gonna pay for it, they're gonna make up for the water, they're gonna do all the stuff to make this not as evil as most of these massive data centers seem to be. Catch us up, what's the headline here? What do we need to know?

Charlie Harger (01:02)
Well, the real

headline is Microsoft realized there's a lot of opposition to these data centers, how much electricity they're sucking up, how much water they're using, and communities are freaking out, right? Because in these communities, they fear their electricity prices are going to go up because there's a data center there, and they're not going to have enough water. So Microsoft actually lost a couple multi-billion dollar proposed projects because of this over the past

year or two with community opposition. the smart move here is to say, listen, we are going to address your fears. And here's how we're going to address them. We will pick up all the ⁓ electricity overages that your community would see. We will make sure that the water that we are turning into steam in order to keep our our servers cool will not be completely going to waste. You don't need to worry about not having enough water to drink because we

you put a data center ⁓ where you live. So Microsoft in a lot of ways is acknowledging reality. The reality is that we are at the basically the bottom level of where AI and energy consumption is. And if even if you look like four or five years from now, I was looking at the numbers for 2030, it's gonna be like at least four times as much electricity used by data centers from where it is now.

It's

using up about 3 to 4 % of all electricity in the United States. So we're talking like one out of every eight kilowatt hours in just four years from now goes straight to AI, nothing else. And so there's a lot that needs to be addressed because of the intense energy needs of AI.

Jason Haworth (02:56)
So there's the small modular reactors. So little tiny small nuclear reactors like you would find inside of an aircraft carrier or something like that or a nuclear submarine. And all the people trying to crank out these AI factories and they're trying to this out. Like this is something that they want to put forward and move that in. And there have been enough advances that this could potentially help bridge some of those gaps. I get that. ⁓

The freshwater part is the part that always kind of gets me nervous. So if you've got large bodies of freshwater, like we do here in the Pacific Northwest, you've got lots of hydroelectric power that can move those pieces across. mean, water in that regard is useful. And then using the water to cool things in those kinds of scenarios kind of makes it an optimal use case. But if you're not going to go with nuclear reactors and you're not going to drag those things across, I mean, the next things that make sense are like wind and solar and...

Burning melted dinosaurs like forever and ever like we just seem to want to do and I don't know maybe invading Venezuela and stealing a other oil so we can power AI or you know going and taking natural resources out of Greenland because you know we need the small precious metals that can only be found in so many areas. Along those lines how the fuck is Microsoft going to make water? Like where is this going to come from that they claim that they're going to help?

Jeremy Grater (03:54)
You

you

Charlie Harger (04:12)
Yeah,

I think your key point of the Pacific Northwest, the northern latitudes in North America, not as much of an issue, right, as if they're to build a data center out in Arizona and Texas, what have you. So those are a lot of considerations. They're really going to, you know, you can make this broad announcement. I'm sure they have a plan. I would really like to see precisely how they plan to do that. And, know, to your point about, you know, small reactors.

Jeremy Grater (04:37)
Yeah.

Charlie Harger (04:42)
You know, Microsoft, what was it last year, just brought up ⁓ a dormant reactor at Three Mile Island. So this is not some sort of sci-fi future speak. So you have ⁓ a lot of atomic companies that are coming aboard. think one is called the Atomic Company. I forget the name of it, but they're really gung-ho. ⁓ And you will also note as well that fusion is becoming more

and more ⁓ prominent in terms of even President Trump. saw this, right? That he folded his publicly traded company into a fusion energy company. I think the Trump family might have a sense that is where the money is because right now we do not make enough electricity at all to keep up with the increasing demand of the robots.

Jason Haworth (05:39)
Well, and we don't have a grid that'll support it. I mean, that's the other thing that people don't realize is that in order for you to get electrons from point A to point B, you have to have a shit ton of copper. And one, we don't have enough copper. Two, there's not enough copper to like make a national grid to actually transport these things. So you have to put the power source close to the thing so you don't have to run as much line. And we're not known for like doing

that well. Like we're we're we were known for you know creating MVPs of electrical transport for long periods of time and we have an aging deteriorating infrastructure with electric cars that people are trying to pull online and AI so we can make you know better cat videos and better social memes but I don't see the value in this for the consumer or for the people. I see the value in this for people looking to stir money into people's pockets.

And the small communities saying, hey, come on in. We'll let you put these places that put these pieces in, just like we used to have. You know, people say we'll put a refinery next to your town and we'll put a power plant or a transfer station in your neighborhood. This is going to be an envy problem. And do you think this is the not in my backyard call of the future where people are going to be like, yeah, I guess you can put in a small cell nuclear reactor or a fusion reactor next to my hometown because we're going to have really good water.

Charlie Harger (07:04)
Yeah,

yeah, I wonder if that is the case because I also don't see in.

alternative, right? That somehow ⁓ AI is just going to stop, that we're going to stop producing it. That's what I can't get my head around. And, you know, I'm glad at least Microsoft is addressing the issue, right? They recognize as a problem. They say, here are the issues. You haven't heard Sam Alton say that. You haven't heard Elon Musk say that. You haven't heard ⁓ any of the Chinese AI companies say that. So maybe we do at least need to be putting a stake in the ground.

out.

Jeremy Grater (07:44)
And I think, you know, that when you look at these big tech companies and the usual lists of like who's the biggest monster on the planet, you Microsoft usually ranks lower, right? They're like 10 or 12 when the big guys are up one or two. So it's not surprising that they would sort of be the first ones to say, hey, we recognize there's a thing we're doing that could be bad. We should do something about it.

but also the environmental impact of AI, the water we're talking about here, it's just gonna dry the planet up faster than anyone's aware of. is this a PR play? Is this them trying to get good headlines and get ⁓ positive attention on this? Or do they recognize a problem? And are they genuinely trying to be in front of a solution?

Charlie Harger (08:22)
I think,

well, first of all, they clearly recognize it's been a billion dollar problem for them when they can't build these data centers and they have the community opposition. So they get that, but it's all PR. It's always PR because that is the lifeblood they have right now. They at least have a little bit of credibility. Compared to Microsoft 30 years ago, where the federal government in the United States had to threaten to split

them apart. were a monopoly and they were doing some not great things. ⁓ I think they recognize that and want to get ahead of the PR now because this is going to be a bigger and bigger issue. at least saying what they believe the solution is, maybe that helps bring other AI companies aboard as well.

Jason Haworth (09:14)
One it-

Jeremy Grater (09:14)
But also the

impact on people too, like how many jobs does this create? Are these data centers just machine, like buildings full of machines that run on autopilot or are there actually gonna be people in there that this will be populated by workers?

Charlie Harger (09:27)
Well, I

imagine that AI plays a big role in that, Jeremy. So I don't want to say exactly how many. Right now, there are going to be a lot of people. Who knows what that'll be 10 years from now.

Jeremy Grater (09:32)
Yes. Sure.

Jason Haworth (09:41)
Someone has to swap hard drives, someone has to pull cable, someone has to run disks, someone has to clean the spills when the water busts of the water cooling system and drops onto things. There's going to be a requirement for meat puppets for a while until we figure out a way to automate these pieces and create little tiny machines to go through and start pulling things out of racks and putting the right cables in the right spot. And along those lines, Microsoft is number four in terms of market cap globally.

So I they're worth three point four to six trillion dollars. The most valuable Nvidia, the people that make the chips after that, it's Google. After that, it's Apple. Number five, Amazon. So you can just look at this and go, huh? So the top five companies all have their hands in AI, consumer electronics and invested interest in building out these big data centers. And there is something definitely to be looked at and feared.

by the fact that these people don't have constraints against them. Like we've deregulated this to the point where we don't want anything getting in the way of these people building these new technologies because we think this is an arms race and this is a geopolitical arms race and if we slow down we're gonna get taken over by somebody else's better AI and the only country that really has a chance of doing this to us is China.

They're not as constrained by environmental factors as we might be, or have not been traditionally. And I think we're becoming more like them than they are like us. So the current administration certainly seems to be more open to that. don't... Charlie, what do you think of that?

Charlie Harger (11:21)
I think

you really have hit the nail on the head, right? Because... ⁓

Do we want to live in a world? This is really West versus East. Do we want to live in a world where the dominant technology is dominated by the Chinese, or would you prefer it be the Americans? Now, I know you could go back and forth. Well, China bad, America good, or America bad, China good. ⁓ But it really does come down to who is going to be the leader in this, who is going to have the, ⁓

you know, the standards with this, who's going to develop those standards with AI? And who do we want to win? Honestly, do you want China to have the better technology or the United States? It's a pretty easy answer for me. And I hate to be so tribal, but it's also like I want the West to be more successful than China.

What do you guys?

Jason Haworth (12:28)
And I don't think that's an unfair statement given where we're at in society. But I don't know that we're not just homogenizing across all aspects of this. Because if you look at the way that China is actually building out their technology stack, their technology stack looks like it's about two to three years behind ours in terms of the strict chip processors and the capability for them to actually go through and do some of these higher level order of operations. But their ability to write software and work around those kinds of problems.

is vastly superior to what we're doing in the US. And I think that's because necessity is the mother of invention. And they're said, and they're told, you know, go figure out a way to make these pieces work. And I think the homogenization of the technology and the way that they're trying to put these things into play is definitely a big issue. I think their motivations and our motivations are not entirely different. And I don't see a lot ⁓ of light between their perspectives as to what they want to do with the technology, because I certainly don't think the current administration

doesn't want a police date. I think they do. I think they want ownership, control and operation of the infosphere. think they want ownership, control and operation of the way people think. And they are not interested in allowing these things to continue. And they're more than willing to wrap it up in red, white and blue and God and country when not really doing anything much different and potentially doing things even worse because they're just essentially going through and brainwashing.

I mean, we- we- we're- they're using the technology to go through and change people's minds to get them to think in two different buckets and camps right now. Like, we are the red versus blue halo version of that that's happening live right now, and it's fucking stupid, and it drives me insane because I don't think anybody is a winner in this.

Charlie Harger (14:12)
I'm also curious.

what your thoughts are on getting back to the energy itself in terms of it almost being a Manhattan Project sort of thing. It's to fusion. Fusion is nowhere near where it needs to be. But that really does become a magic bullet in a lot of ways in terms of reducing carbon emissions, in terms of being a clean energy that you don't have to worry about to at least any of the extent of fission. ⁓

Jason Haworth (14:24)
Mm-hmm.

Charlie Harger (14:44)
That is something that would have all sorts of carry on effects, not just with AI, but in terms of having a realistic ⁓ future for a growing power grid, in terms of being able to use EVs in a expanding way. What are thoughts on that?

Jason Haworth (15:03)
right?

So ⁓ I think you can see what the geopolitical interest is right now in regards to this because the US, China and several other countries all have moon projects going on. And the reason they have moon projects going on is because the largest source of material for fusion is helium-3. And there's a shit ton of it on the moon and we're willing to reinvest and go to outer space to get our electricity and pull it back down again to make these pieces work. I think it's huge and I think

You can go even further because I think it was Bezos that came out and said, I can see in the not too distant future, giant data centers in space that are powered by solar radiation and are cooled by space technologies, whatever the fuck that is, and that energy being beamed down to the planet.

It does feel like science fiction that these guys are looking at figuring out ways to go through and steal electrons from the moon or create a giant floating space laser that's gonna go through and burn up the planet. Exactly, right, like it's freaking laser beams on sharks' heads. this is crazy. We've created a bunch of Dr. Evils, and I'm not calling them evil. Like, I don't know. I mean, some of them probably fucking are. But at the end of the day,

Charlie Harger (16:09)
Ha ha ha!

Jeremy Grater (16:10)
Hehehehehe

Jason Haworth (16:22)
That's not the thing. Like the thing is, is that we're trying to make advancements to our society in an interesting way. And these decisions affect everybody on the planet. It's not just about making people richer. And we're subsuming or we're giving up a lot of our agency as individuals, as people, and as a collective to these large corporations because we think we're going to get something really good out of it, which we might, but we also might burn up the planet and the process or even worse.

create our future robot overlords that make little tiny machines and microviruses that turn us all into their robots to do things. I mean, the possibilities are not limitless, but they're a lot more vast than I think people realize.

Jeremy Grater (17:03)
Something I think is interesting about all of this, mean, just the crazy links we're talking about, about going to the moon to harvest energy and all these things. I also see as many articles every day of, the AI bust is right around the corner. So all of this is going to just blow up in our faces anyways. I don't think the amount of money that Microsoft, even with the story we started talking about with these data centers, the amount of money and energy and power that it's going to take to make that happen, add to that moon mining and all of the things.

None of that indicates to me that this is something that's going away anytime soon. And I don't see this being something that's going to financially bust. I might eat those words in six months or a year. Who knows? But I see a lot of people throwing a lot of money at this thing, which indicates to me that this is not going anywhere. And so any any time I see a company like Microsoft that's doing something that at least on the on the surface looks like they're trying to do for the right reason, I'm going to get behind it because there's a lot of evil that can power something like this for a lot of bad intentions.

Charlie Harger (18:00)
As

⁓ an old guy now, just turned 50 Jeremy, just turned 50.

Jeremy Grater (18:05)
I'm right behind you, brother. I'm right behind you. You're the oldest one here, grandpa.

Jason Haworth (18:06)
Wait a minute, I'm the oldest one on the call? ⁓ damn it!

Charlie Harger (18:11)
lived through the dot com bust of the late 90s. I worked for a startup, we thought we were all going to be millionaires by the time we were 30. And ⁓ dot com had a bust. Internet did not bust the underlying technology did not bust some of the harebrained business ideas that that were using the technology. Those are the ones that went bust. So AI

Jason Haworth (18:37)
100x valuations.

Jeremy Grater (18:39)
You

Charlie Harger (18:40)
Those stock

options, they were beautiful.

Jeremy Grater (18:44)
Hahaha!

Jason Haworth (18:44)
I worked for Altium, and got bought by Nortel

for 70 to 1, so I was on one of those rides and then it exploded.

Charlie Harger (18:50)
Yeah,

yeah. So I guess my thinking is that there is no turning back on AI. Some of the AI companies certainly are going to go. But the underlying technology, I don't think there's turning back.

Jeremy Grater (19:07)
Absolutely.

Jason Haworth (19:08)
Right. Well,

and you can look at this from a different angle right now to make AI work and to make all the infrastructure work and everything else. You require humans. You require people to get up and do things. So there's jobs that are going to be available for a long period of time until we create robots that can actually do these human like things. And if none of you have watched it, I highly encourage you to watch all the general dynamics pieces, all the general robotics pieces where they're making

things that do human things well and better than humans. I would also encourage you to reach out and watch the other ones where they have made robots that are supposed to do human-like things that freak out and attack humans. Those are all worth watching.

Jeremy Grater (19:51)
My favorite video, my favorite video on the

internet right now, is a guy who's like doing a martial arts technique and it's like mo-capped to the robot, the robot's mimicking him. They turn around, the robot kicks him square in the nuts. It's my favorite thing to watch on the internet. It is the best.

Jason Haworth (19:59)
Yup.

FWAH!

Charlie Harger (20:02)
Yeah

Jason Haworth (20:05)
This is just your America's Funniest Home Videos pitch like all over again.

The show, Getting Hit in the Nuts is like the perfect title. Now it's getting hit in the nuts by robots. it's so good.

Jeremy Grater (20:12)
Yes.

The best, the best.

Charlie Harger (20:16)
I think

you

Jeremy Grater (20:17)
Charlie, I know ⁓ this is a pretty big deal for a podcast. If you want to promote your little radio show to try and grow an audience, feel free to mention it.

Charlie Harger (20:23)
I'm sorry.

Jason Haworth (20:24)
Ha ha ha!

Charlie Harger (20:26)
Well, thanks for your opportunity, Jeremy. No, you can. I host Seattle's morning news, KIRO News Radio, mynorthwest.com, every weekday morning. Check it out. I'm on video. I make funny faces.

Jeremy Grater (20:30)
Hahahahah

He does make funny faces. Charlie and Amanda, it is a fantastic show. Charlie and I worked together for long time in radio. He's one of the best guys in the business. A real, I worked with a lot of people in this business. I wouldn't call them all journalists the way that I call you a true, a real newsman, a journalist. I respect the hell out of you and the work you do. Thanks so much for making a few minutes for us on our little project here.

Jason Haworth (20:41)
Fantastic show, folks.

Charlie Harger (21:02)
You bet I'm off to do

my investigative report. There's this company called Skynet and it sounds like they have some fantastic products.

Jeremy Grater (21:06)
Never heard of them.

Jason Haworth (21:10)
I would definitely invest heavily in Skynet for the short term. And yeah, maybe go short on human beings. Thanks, Charlie. No

Jeremy Grater (21:17)
Thanks, Charlie.

Charlie Harger (21:17)
Take care, guys.

Jeremy Grater (21:21)
right, that was Charlie Harger, one of the hosts of Seattle's Morning News on K-I-R-O, KIRO Radio in Seattle. You can check their website out. It's mynorthwest.com. That's where we found this article. And you can find out more about Charlie and his show there on the radio and where you can listen to it, whether you're in the Seattle area or not.

And of course, we'll have that linked at our website, probots.me, we'll be back Monday morning with another episode. Thanks so much for listening. think this would be helpful for someone to listen to, feel free to pass it along. Again, those links are at probots.me. We'll see you next

 

Charlie Harger Profile Photo

Host, Seattle's Morning News

I’m a journalist with more than 30 years of experience covering the stories that shape life in western Washington. I host Seattle’s Morning News on KIRO Newsradio, where I help listeners start their day informed, prepared, and sometimes even entertained.

My reporting has earned multiple national and regional awards, including several Edward R. Murrow Awards. I’ve also been named APTRA’s Radio Reporter of the Year and recognized for investigative reporting on major stories like the Boeing 737 MAX crisis and interviews with serial killer Gary Ridgway.

I specialize in making complex issues understandable, focusing on how news affects real people. I’m a fourth-generation western Washington resident and have called Auburn home for over a decade.

Before hosting, I served as News Director at KIRO, worked at KOMO for nearly two decades, and taught journalism at Green River College. I take news seriously — but believe the best reporting connects on a human level.