Exposing The Dark Side of America's AI Data Center Explosion — View From Above — Business Insider

[00:00:01] Speaker 1: Giant warehouses are popping up across the U.S. more than two every week. They feed AI algorithms, store photos and answer our questions. [00:00:11] Speaker 2: Hey Google. [00:00:12] Speaker 1: A data center campus like this can consume as much power and water as an entire city. And many of the biggest server farms are emerging from the desert. But there's no official record of how many of these are being built, where they are or even who owns them. [00:00:29] Speaker 2: To be honest, I've never really run into so much resistance for records than this project. [00:00:35] Speaker 1: Big tech companies often go to great lengths to hide the details. So we decided to build a map. [00:00:43] Speaker 2: What we designed is a system to see not only the individual location of each data center across the country, but also where they cluster. [00:00:53] Speaker 1: And for some, these clusters are appearing too close to home. We went to meet people who say having data centers next door is affecting their health. [00:01:04] Speaker 3: We had a hard time breathing. I couldn't sleep. I was like, I thought I was losing my mind. [00:01:09] Speaker 4: Summertime is more of a... During the winter, it's like... Almost like an engine trying to start. [00:01:19] Speaker 1: So how many data centers are there in the US? And is there enough power and water to satisfy a building boom that is only just getting started? Where I am standing now is at the middle of the data center capital of the world. This is Loudoun County, Virginia, otherwise known as data center alley. And it may not look like much from down here, but just take a look from up above. These are Amazon data centers. Zoom out and you can see tech companies tend to build data centers in clusters. Where there is a reliable power supply, access to enough water, as well as tax breaks and affordable land. But of course, these tend to also be places where lots of people already live. Like Santa Clara County, California. Or the most heavily populated part of Arizona, Maricopa County. And of course, here in Northern Virginia. The most densely populated part of the Washington DC metro area. And one of the largest and fastest growing residential areas in the US. [00:02:42] Speaker 3: Yeah, I mean, I was the fourth homeowner in this neighborhood. Fourth to occupy a house in this neighborhood. [00:02:51] Speaker 1: Donna Gallant has lived on this street in Prince William County, Northern Virginia, for the last 30 years. All very peaceful until 2021, when things started to change. These are Google data centers. And ever since they started rising from the ground, Donna has been looking for answers. [00:03:10] Speaker 3: There's no transparency. [00:03:12] Speaker 1: Both. [00:03:13] Speaker 3: So anytime you ask a question, it's, oh, we signed an NDA, we can't talk about it. Oh, we're under NDA, we can't talk about it. So there's no transparency at all. [00:03:25] Speaker 1: The site isn't complete yet, but she says the noise is already taking its toll. [00:03:32] Speaker 3: When I go to my room at night, the tonal noise immediately triggered my anxiety. To the point where I couldn't sleep. I ended up having to go down on my first floor and put noise cancelling headphones on just so I could sleep. [00:03:54] Speaker 1: But for Donna and her neighbors, this is just the start. Over the next few years, they will find themselves surrounded by a string of new data centers. [00:04:06] Speaker 3: That's going to go from this road here all the way and around my neighborhood. That's where the data center is going to go. And that's going to be 75 feet tall, dead smack in the middle of a neighborhood. [00:04:21] Speaker 1: Zoom out a little more and you can see this plot of land. It was originally meant for housing, but in 2023 it was rezoned to clear the way for the construction of data centers instead. Prince William County already has over 70 data centers, but if this master plan for a data center opportunity zone is fully realized, together with Lowndon County to the north, they'll have more data centers than Russia. Dona set up a local campaign to challenge the rezoning, but the lawsuit was dismissed. [00:04:58] Speaker 3: Yeah, it's heart-wrenching. It's heart-wrenching. Do I have ill feelings towards the data center industry for that? Yes, I do. I think we all need it. It's the way the world is going, but do I think dropping it next to people's homes is the right answer? No. [00:05:19] Speaker 1: There is no definitive public directory of data centers, no official map, no single regulator to ask or government agency to foyer. [00:05:28] Speaker 2: So, you know, it's been really tricky to kind of get these records because the companies don't want to disclose all of that information. [00:05:36] Speaker 1: By tricky, we mean redacted records and requests denied on the grounds of trade secrets. It turns out there is one thing that most data centers need: backup generators, in case the grid fails. And anyone who wants to install a generator needs to apply for an air quality permit. [00:05:57] Speaker 2: So what we set out to do was request all of the permits that are issued to data centers for those backup generators. [00:06:05] Speaker 1: That meant filing public record requests for air permits in every state. They list the capacity of the generator so we can extrapolate the power needs of the data center. They also provide clues about who owns it. Take this hotspot rapidly expanding close to Columbus, Ohio. [00:06:28] Speaker 2: There are at least 164 emergency generators permitted here. This is a data center where the air permit was applied for by an LLC called Magellan Enterprises LLC. But the company has applied for a trade secrets exemption. So they were actually able to redact pretty much all the information that we wanted. [00:06:51] Speaker 1: But all of the big tech giants must disclose any companies like LLCs that they own. And by digging into their official records, we managed to pull back the mask. And it turns out that the data center in Ohio. [00:07:07] Speaker 2: It's not Magellan Enterprises LLC that's actually owned by Google. [00:07:12] Speaker 1: This is a task we repeated hundreds of times to build this map. It's the most comprehensive tally to date of America's exploding data center industry. Every dot is a data center large enough to need a permit for its backup generators. These are the facilities either already built or approved for construction at the end of 2024. We reached a total of 1240 data centers. That's nearly four times the number in 2010. [00:07:41] Speaker 2: The companies that use the most amount of power in data centers across the country, it probably won't surprise you, are Amazon, Microsoft, Google, Meta and QTS. [00:07:51] Speaker 1: Of the data centers we pinpointed, 177 belong to Amazon. And we've just passed the huge Amazon data centers behind us and already we're straight into a residential area. And you can start to see why these data centers are so controversial. Those boxes on the roof there, that's the ventilation for the cooling systems to keep those hundreds and hundreds of servers from overheating and overloading. And these are running 24/7. And there's nowhere else in the world with a higher density of data centers than here in Loudoun County, Virginia. As much as a third of the planet's internet traffic flows through the state of Virginia, the 329 data centers we tracked, together consumed almost a quarter of the state's electricity in 2023. Tight security by the looks of it, we've got pretty high fences with barbed wire on top, I don't think we can get much closer. It's rare to get a glimpse inside one of these server warehouses. We contacted dozens, asking for a tour. Eventually, a small company called Lunavi in Wyoming said yes. This 35,000 square foot data center is tiny compared to the largest ones in the US. Those can stretch to well over a million square feet. Lunavi offers cloud services to customers like betting apps, mapping companies, insurance and healthcare businesses. And if you look at that, we can see that the location is no accident. Cheyenne, the capital of Wyoming, sits on an east-west internet superhighway. The city is also well connected to renewable energy, generated by these wind farms. Wyoming is only a small player compared to Virginia. And to attract more, the state is offering generous tax breaks to encourage the big tech giants, like Microsoft. This large data center began emerging from the map south of Cheyenne in 2021. And this is a site belonging to Meta, just over the road. [00:10:17] Speaker 5: You're currently within the critical infrastructure space of data center 2. Designed to have 800 cabinets, roughly, at full capacity on that side. So we take the air from the above ceiling grid. We feed it down through the computers. They heat the air up, comes back out the back and up through the chimney above ceiling. And then just continue to circulate the air on that perspective. [00:10:41] Speaker 1: Lunavi has the space to scale up to the more energy-hungry GPU processors used to train AI algorithms. [00:10:48] Speaker 6: So this big white space is future computing. The cabinets in this facility now have a capacity of something like 5 to 10 kW. If they want to build the ones that are going to do AI with graphic processing units, GPUs, that'll go up to something like 70 to 100 kW. [00:11:07] Speaker 1: GPUs do need more power, but they can complete more tasks than a regular processing chip and in less time. The lower ambient temperatures in the high plains of Wyoming mean data centers use less energy and water to cool themselves. Lunavi say they use around 500,000 gallons of water a month when at full capacity. That's roughly the amount of water used by 200 people. [00:11:31] Speaker 5: The nice cool ambient air temperatures allow us to maintain a low PUE, which is a power usage efficiency, which then allows us to just be a lower cost center for our customers and pass those savings on. [00:11:46] Speaker 1: But this area has been plagued by droughts. So farmers are watching the data center campuses emerging to the south of Cheyenne and getting nervous. This is a story unfolding not just in Wyoming and in Virginia, but across the country. The average American holds hundreds of gigabytes of data in the cloud. Big tech companies, the healthcare and finance sectors, governments, of course, hold much more. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:12:20] Speaker ?: The average American has been in the cloud. [00:12:20] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:12:29] Speaker ?: The average American has been in the cloud. [00:12:30] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:13:01] Speaker ?: The average American has been in the cloud. [00:13:02] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:01] Speaker ?: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:05] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:10] Speaker ?: The average American has been in the cloud. [00:14:11] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:25] Speaker ?: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:29] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:36] Speaker 4: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:14:40] Speaker ?: The average American has been in the cloud. The average American has been in the cloud. [00:14:42] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. [00:14:44] Speaker 4: The average American has been in the cloud. [00:14:55] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:15:03] Speaker 4: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:15:13] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. [00:15:25] Speaker 4: The average American has been in the cloud. [00:15:26] Speaker ?: The average American has been in the cloud. [00:15:26] Speaker 4: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:15:34] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:15:55] Speaker 4: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:16:08] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:17:06] Speaker ?: The average American has been in the cloud. The average American has been in the cloud. [00:17:08] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:17:12] Speaker ?: The average American has been in the cloud. [00:17:13] Speaker 4: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:17:35] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:17:55] Speaker 7: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:18:03] Speaker ?: The average American has been in the cloud. [00:18:04] Speaker 7: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:18:07] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:18:29] Speaker ?: The average American has been in the cloud. [00:18:30] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:18:55] Speaker 7: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:19:06] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:19:25] Speaker 2: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:19:37] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:20:01] Speaker 2: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:20:25] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:20:41] Speaker ?: The average American has been in the cloud. [00:20:42] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:20:55] Speaker 2: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:21:03] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:21:24] Speaker ?: The average American has been in the cloud. [00:21:25] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:22:44] Speaker 7: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:22:48] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:23:14] Speaker 2: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:23:25] Speaker ?: The average American has been in the cloud. [00:23:26] Speaker 1: The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. The average American has been in the cloud. [00:23:44] Speaker 2: And it's a magnitude that we've never really seen across the country before. It's really startling. [00:23:48] Speaker 1: As states race to fuel the AI boom, some are reversing their green energy promises and turning back to power from coal and natural gas. In Nebraska, the two largest electricity utility companies committed several years ago to net zero emissions from electricity generation by 2050. But just this single metadata campus in Springfield, Nebraska, could use as much power in a year as 400,000 homes. One of the state's largest public utilities voted to postpone closing down two coal-fired power plants here in Omaha. This image, created using satellite data in June 2023, shows a plume of CO2 emissions released from the North Omaha Station facility at an estimated rate of 300,000 kilograms per hour. And in 2025, the utility decided to build two new natural gas plants, all to meet spikes in electricity demand, mainly driven by data centers. And let's not forget the thousands of backup generators we tracked, which, even if they run for just a few hours a month, spew harmful pollutants into the air. [00:24:58] Speaker 2: That means that utility providers are either abandoning their commitments or significantly stalling their commitments to move away from fossil fuel. They're not going towards renewable energy sources, such as solar or wind, for example, because those resources right now can't sustain the massive power consumption demand. [00:25:20] Speaker 1: Big data center developers have announced massive investments in renewable projects like solar plants, wind farms or nuclear power. For example, in Pennsylvania, Microsoft has struck a deal to buy power from the notorious Three Mile Island nuclear power plant when it reopens in 2027. The plant suffered a partial meltdown in 1979. But just like with the water credits, big tech companies are also looking to offset their carbon footprint by paying others to deliver renewable energy to the grid on their behalf. The big question is, can the already fragile and fragmented grid support it? And if massive infrastructure upgrades are needed, who picks up the tab? Amazon, Microsoft and Google told Business Insider they were committed to paying their full share for upgrades to grid infrastructure, like high voltage power lines. But there's plenty of evidence pointing to the fact that costs are already being passed on to customers. In Virginia, Dominion Energy disclosed that it would need to roughly double its electricity generation by 2039, mainly to meet demand from data centers and electric vehicles. The expansion could cost up to $103 billion, increasing residential electricity bills by as much as 50%. Despite the burden on the grid and on water supplies, some states are doing all they can to attract data centers into their backyards. Many regions simply don't want to miss out on the AI boom or the kudos of having a big tech employer on their doorstep. So far, though, the promise of large numbers of jobs has not materialized. A Business Insider analysis found that even the largest data centers employ fewer than 150 permanent workers, and some have as few as 25. But the tax breaks keep on coming. [00:27:13] Speaker 8: What we found is that massive corporations are benefiting from various taxes in different states for their data centers. [00:27:26] Speaker 1: We tracked 37 states offering tax incentive programs to data centers, like zero tax on building materials, machinery or equipment, but also preferential rates on water and electricity. In Virginia, 56 data centers received tax savings of almost $1 billion in the 2023 fiscal year alone. Take a look at these data centers in New Albany, Ohio. [00:27:52] Speaker 8: In 2017, a little known LLC called SideCat went to the city of New Albany in Ohio and said we are going to build two gigantic data centers on about 300 acres of land. In return, could you please give us 100% property tax abatements for at least 15 years? And the county said, okay. By our estimate, SideCat LLC received at least $60 million in fork on taxes. And then, you know, everybody found out that SideCat LLC is not a mom and pop, a data center company, but was Meta, the parent company of Facebook. [00:28:33] Speaker 1: The rollout of AI across industries, from social media to medical care, will accelerate data center construction. In 2025 alone, Meta planned to spend at least $64 billion on facilities and equipment, Google $75 billion and Microsoft $80 billion. And the ultra powerful computer chips driving the AI tools will consume more and more energy. The 2024 Department of Energy report estimates that their electricity use, driven by the AI boom, could soon require as much as 12% of total US electricity use. In 2023, it was just over 4%. For those like Donna living alongside these data farms, there are decisions to be made. [00:29:24] Speaker 3: I don't think a lot of my neighbors truly understand the severity of what's to come. And I'm hoping to be out before that happens. Breaks my heart. But I can only fight for so long. And then I'm going to wave the white flag and I'm going to pack up and I'm going to leave. I'm going to wave. [00:29:44] Speaker ?: I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the white flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag. I'm going to wave the flag.