Guest Post by Peter Reagan
In recent months I’ve been asked by numerous readers to give my take on the AI data center issue and the debate it has ignited across the U.S. In response, I’ve jumped into an interesting and steep learning curve, but it seems to parallel information I’ve covered in previous articles on AI.
From what I can tell, the politics of the data center debate don’t break cleanly along party lines. There are Democrats and Republicans on both sides, though their reasons for opposing the centers are not necessarily the same. The Trump Administration has firmly embraced accelerated AI and data center development, framing it as an issue of economic competitiveness and national security. Even so, some of the strongest resistance is coming from deeply conservative communities.
I call it “infrastructure,” but I have to say, one of my biggest apprehensions about data centers is the mismatch between their enormous physical footprint and the benefits that flow directly to local communities. They certainly create construction work, permanent jobs and tax revenue. But these are extraordinarily capital- and energy-intensive facilities, and I’m struggling to see why the benefits should automatically outweigh the local costs.
I’ll try to break it down in basic terms…
Much of the current boom in new hyperscale data centers is being driven by AI. These facilities provide the enormous computing capacity required for two basic jobs: Training AI models, and running those models once they’re deployed, a process called “inference.”
The most ambitious AI campuses contain enormous clusters of specialized chips used to train so-called “frontier” models. Training is computationally intensive and can run for long periods at very high utilization, which is one reason these facilities consume so much electricity. Inference – actually serving all those AI requests after a model has been trained – requires substantial computing capacity of its own.
And the computing equipment doesn’t become obsolete once a single model is finished. The same infrastructure can be used for additional training runs, new generations of models and the growing workload required to serve existing AI systems.
We can see and use cars, we can see and use televisions and refrigerators, we can see the workers on the factory floor and we can see the jobs these products create. Not so much with AI and data centers, and this is one of the problems with trying to convince the public that such operations are necessary for the future well-being of the country.
In some cases AI training facilities convert partially or fully to inference operations. Meaning they handle the day-to-day functions of existing AI. So, every time you ask ChatGPT a question, or ask it to create an image for you, or ask it to make a ridiculous, uncanny valley video for you, that request is processed by a data center somewhere.
All of that computing uses electricity – a lot of it. A 2026 update from Lawrence Berkeley National Laboratory estimates that data centers could consume about 11.8% of all U.S. electricity by 2030, with plausible scenarios ranging from 9.5% to 15.3%. AI isn’t solely responsible for that demand, but it is one of the major forces driving the increase.
If this all seems highly centralized, that’s because it is. There are also legitimate privacy questions surrounding cloud-based AI. Depending on the service and your settings, conversations may be retained and some may be used to improve future models. Both ChatGPT and Gemini now offer data-use controls that limit how conversations are used. So maybe every interaction doesn’t automatically becomes training material. Still, using a remotely hosted AI service necessarily means entrusting that information to someone else’s infrastructure, security and internal policies.
In other words data centers exist to grow more and more AI models while making them smarter and smarter (in theory). But again, why should the average American give a damn about this?
Governments and industry leaders certainly don’t talk about AI as if it’s just another software upgrade. The Trump Administration’s own AI Action Plan describes leadership in the technology as critical to economic competitiveness and national security, and calls for rapid construction of the data centers, energy infrastructure and semiconductor facilities needed to support it.
The idea is that AI will run almost everything and that our entire evolution as a species will be transcendent because of AI. Some of AI’s strongest advocates go considerably further, predicting artificial general intelligence and eventually systems that surpass human capabilities across a broad range of tasks. There is a serious debate over whether, when or even if those milestones will arrive.
Personally, I’m skeptical of the more utopian predictions. I don’t think we need a science-fiction scenario to find reasons for caution. AI is already powerful enough to change how people work, research, communicate and make decisions. Those consequences are worth examining without assuming a machine will suddenly become all-knowing.
I don’t think the singularity is a reality – I think it’s a pipe dream. That said, I do think AI creates a very real temptation to outsource more of our thinking than we should. As I’ve said in the past, I’m less worried about the rise of some “machine god” than I am about ordinary people becoming so dependent on the software that they stop checking its work, doing their own research or exercising their own judgment.
The “information economy” is a more grounded notion of how AI might be applied, but it’s still a bit ridiculous. It’s the idea that larger economies will no longer rely on the production of tangible goods and resources to support the system. Rather, they will rely on the creation of digital products and the trading of data derived from algorithms.
There’s also a more grounded concern: What happens if companies build far more AI infrastructure than customers ultimately need or are willing to pay for?
Reuters recently compared today’s data center construction frenzy with the broadband boom around the turn of the century. Enormous amounts of fiber infrastructure were built in anticipation of future demand, only for some of those assumptions to prove wildly optimistic. AI may ultimately justify today’s buildout – but given the sums involved, the possibility of overbuilding shouldn’t simply be dismissed.
The Trump Administration has been quite explicit about why it supports this buildout: It views AI as an industrial and geopolitical competition, particularly with China. Its AI Action Plan ties American leadership in AI to economic prosperity, national security and scientific leadership, while calling for faster permitting of data centers and the energy infrastructure needed to power them.
I can’t speak to the military applications because, frankly, they were entirely theoretical until recently. Now, the Pentagon is using AI for tasks ranging from intelligence analysis and logistics to battlefield threat identification. There are plans to expand AI’s role in American armed forces. What remains uncertain is how beneficial today’s AI models will ultimately prove to be in warfare – and whether the enormous civilian data-center buildout now underway is necessary to secure that advantage.
The competition with China isn’t imaginary. Both countries are pouring resources into advanced AI, and Washington now openly treats maintaining U.S. leadership as a strategic priority. That doesn’t automatically mean every data center project is justified, but it does explain why federal officials are reluctant to slow the overall buildout.
So far, much of the opposition is rooted in practical local concerns: Electricity demand, utility bills, water consumption, loss of farmland and the effect enormous industrial facilities can have on quality of life nearby. In other words, these aren’t abstract objections to AI. They’re complaints about what happens when a massive new industrial neighbor arrives.
Increasingly, developers are looking beyond major cities for places with large tracts of affordable land and, above all, access to enormous amounts of power. Renewable generation can be attractive, but so can natural gas, nuclear power or simply an existing grid connection with spare capacity. In the AI data center business, access to electricity is rapidly becoming the deciding factor.
As someone who lives in one of these communities, I can tell you the sales pitch isn’t landing very well where I live. It’s a heavily conservative area, but support for Trump doesn’t automatically translate into enthusiasm for a giant server farm moving in next door.
And apparently my neighbors aren’t unusual. Opposition to new data centers now crosses party lines and extends through rural, suburban and urban communities.
Maybe if we were adding reliable power generation – specifically nuclear plants – as quickly as we’re adding electricity demand, people would be somewhat less inclined to complain. Tech companies themselves seem to recognize the problem. Several have already signed long-term nuclear-power agreements to help supply their data centers.
AI centers can create thousands of jobs while they’re being built. The permanent workforce is much smaller. Reuters recently reported that an individual Meta data center typically employs around 100 people long-term, although the number varies based on the size and type of facility.
Now, those can be good jobs – technicians, electricians, engineers, HVAC specialists and security workers, for example. But for a project consuming hundreds of megawatts of electricity and billions of dollars in capital, the permanent head count looks surprisingly small.
There’s user access for the country at large. Many people use Gemini or Grok to do nothing more than making memes. Of course, AI can do far more. Businesses are already using it for software, research, customer service and countless other tasks. However, Deloitte tells us the benefits for businesses have been slow to materialize and hard to measure.
My question is different: How much of that economic benefit flows back to the community hosting the infrastructure?
Tax revenue is probably the strongest argument data-center developers can offer local communities, and in some places it has been substantial. Construction work and permanent jobs matter, too. The harder question is whether those benefits compensate residents for the added demands on electricity, water, land and infrastructure – especially when tax incentives are part of the deal.
I’m not completely dismissing the people who want to promote AI development, but none of them are conjuring up any selling points practical enough or exciting enough to set imaginations aflame. It’s not like the space race of the cold war era.
No one is clutching their pearls in fear of “losing an AI race” that may or may not matter. Promises of an AI-driven productivity revolution are much harder for a town to weigh than the immediate realities of a new industrial facility, higher electricity demand and hundreds of acres of development. Future national benefits may be enormous. But local residents are understandably asking what they receive today.
When I look at the AI debate, I see multiple factions, interests and visions colliding. Plenty of people promoting data centers sincerely believe AI will produce enormous economic and technological benefits. Plenty of people opposing them aren’t anti-technology – they’re homeowners, farmers and ratepayers with legitimate questions about land, water, electricity and costs.
That distinction matters. This debate gets a lot easier to understand once we stop assuming everyone on the other side has sinister motives.
There is another problem, though: The incentives aren’t evenly distributed. Technology companies have enormous sums invested in accelerating AI development. Federal officials see a strategic competition they don’t want America to lose. State and local governments see potential tax revenue.
The people living next door to these facilities have a different set of incentives. They’re thinking about their utility bill, their water supply, their property and the character of their community. None of those groups has to be evil or stupid for their interests to collide.
It’s hard to say where this goes from here, but public opinion has moved sharply against local data center construction. An NBC News Decision Desk poll released in September found 69% of Americans opposed an AI data center being built in their local area. A separate University of Pennsylvania survey found opposition rising from 49% early this year to 61% by summer.
Public opinion can change, of course. But trying to force these projects into unwilling communities without addressing their concerns about power, water, land and local costs seems far more likely to deepen the backlash than solve it.
And that’s really the larger point.
Maybe the AI buildout delivers everything its advocates promise. Maybe the benefits eventually justify the enormous investments in land, power and infrastructure. Or maybe we discover (exactly as we have during previous technology booms) that enthusiastic promises got way ahead of reality.
I don’t know which way this goes. What I do know is that ordinary Americans have very little control over the outcome – even though they’ll ultimately share in the costs, whether or not they also share in the benefits.
That’s one reason I believe diversification matters. When so much of the economy is being reshaped around expensive technologies with uncertain payoffs, it makes sense to consider keeping a portion of your savings in something tangible that doesn’t depend on an AI forecast, a data center or somebody else’s business model.
The dollar buys less today than it did a year ago, a decade ago, a generation ago. That slide is the predictable result of endless printing and borrowing. Dr. Ron Paul believes the FED has the answer. But they’re keeping it to themselves. Fortunately, a new financial chart has publicly exposed their secret moves. See it for yourself: https://freekit.birchgold.com/lf/ro...D_v02a_article&placement=article&cid=rp_media
To learn more about how physical gold could help protect your retirement portfolio, click here to get your FREE info kit on Gold IRAs from Birch Gold Group. And now introducing a Crypto IRA to capitalize on the fastest growing market in the world.
Tweet
Continue reading...
[ H/T The Burning Platform ]
In recent months I’ve been asked by numerous readers to give my take on the AI data center issue and the debate it has ignited across the U.S. In response, I’ve jumped into an interesting and steep learning curve, but it seems to parallel information I’ve covered in previous articles on AI.
From what I can tell, the politics of the data center debate don’t break cleanly along party lines. There are Democrats and Republicans on both sides, though their reasons for opposing the centers are not necessarily the same. The Trump Administration has firmly embraced accelerated AI and data center development, framing it as an issue of economic competitiveness and national security. Even so, some of the strongest resistance is coming from deeply conservative communities.
I call it “infrastructure,” but I have to say, one of my biggest apprehensions about data centers is the mismatch between their enormous physical footprint and the benefits that flow directly to local communities. They certainly create construction work, permanent jobs and tax revenue. But these are extraordinarily capital- and energy-intensive facilities, and I’m struggling to see why the benefits should automatically outweigh the local costs.
I’ll try to break it down in basic terms…
What do data centers actually do?
Much of the current boom in new hyperscale data centers is being driven by AI. These facilities provide the enormous computing capacity required for two basic jobs: Training AI models, and running those models once they’re deployed, a process called “inference.”
The most ambitious AI campuses contain enormous clusters of specialized chips used to train so-called “frontier” models. Training is computationally intensive and can run for long periods at very high utilization, which is one reason these facilities consume so much electricity. Inference – actually serving all those AI requests after a model has been trained – requires substantial computing capacity of its own.
And the computing equipment doesn’t become obsolete once a single model is finished. The same infrastructure can be used for additional training runs, new generations of models and the growing workload required to serve existing AI systems.
We can see and use cars, we can see and use televisions and refrigerators, we can see the workers on the factory floor and we can see the jobs these products create. Not so much with AI and data centers, and this is one of the problems with trying to convince the public that such operations are necessary for the future well-being of the country.
In some cases AI training facilities convert partially or fully to inference operations. Meaning they handle the day-to-day functions of existing AI. So, every time you ask ChatGPT a question, or ask it to create an image for you, or ask it to make a ridiculous, uncanny valley video for you, that request is processed by a data center somewhere.
All of that computing uses electricity – a lot of it. A 2026 update from Lawrence Berkeley National Laboratory estimates that data centers could consume about 11.8% of all U.S. electricity by 2030, with plausible scenarios ranging from 9.5% to 15.3%. AI isn’t solely responsible for that demand, but it is one of the major forces driving the increase.
If this all seems highly centralized, that’s because it is. There are also legitimate privacy questions surrounding cloud-based AI. Depending on the service and your settings, conversations may be retained and some may be used to improve future models. Both ChatGPT and Gemini now offer data-use controls that limit how conversations are used. So maybe every interaction doesn’t automatically becomes training material. Still, using a remotely hosted AI service necessarily means entrusting that information to someone else’s infrastructure, security and internal policies.
In other words data centers exist to grow more and more AI models while making them smarter and smarter (in theory). But again, why should the average American give a damn about this?
The global race for AI supremacy and the information economy
Governments and industry leaders certainly don’t talk about AI as if it’s just another software upgrade. The Trump Administration’s own AI Action Plan describes leadership in the technology as critical to economic competitiveness and national security, and calls for rapid construction of the data centers, energy infrastructure and semiconductor facilities needed to support it.
The idea is that AI will run almost everything and that our entire evolution as a species will be transcendent because of AI. Some of AI’s strongest advocates go considerably further, predicting artificial general intelligence and eventually systems that surpass human capabilities across a broad range of tasks. There is a serious debate over whether, when or even if those milestones will arrive.
Personally, I’m skeptical of the more utopian predictions. I don’t think we need a science-fiction scenario to find reasons for caution. AI is already powerful enough to change how people work, research, communicate and make decisions. Those consequences are worth examining without assuming a machine will suddenly become all-knowing.
I don’t think the singularity is a reality – I think it’s a pipe dream. That said, I do think AI creates a very real temptation to outsource more of our thinking than we should. As I’ve said in the past, I’m less worried about the rise of some “machine god” than I am about ordinary people becoming so dependent on the software that they stop checking its work, doing their own research or exercising their own judgment.
The “information economy” is a more grounded notion of how AI might be applied, but it’s still a bit ridiculous. It’s the idea that larger economies will no longer rely on the production of tangible goods and resources to support the system. Rather, they will rely on the creation of digital products and the trading of data derived from algorithms.
There’s also a more grounded concern: What happens if companies build far more AI infrastructure than customers ultimately need or are willing to pay for?
Reuters recently compared today’s data center construction frenzy with the broadband boom around the turn of the century. Enormous amounts of fiber infrastructure were built in anticipation of future demand, only for some of those assumptions to prove wildly optimistic. AI may ultimately justify today’s buildout – but given the sums involved, the possibility of overbuilding shouldn’t simply be dismissed.
What do local communities get out of the “fourth industrial revolution”?
The Trump Administration has been quite explicit about why it supports this buildout: It views AI as an industrial and geopolitical competition, particularly with China. Its AI Action Plan ties American leadership in AI to economic prosperity, national security and scientific leadership, while calling for faster permitting of data centers and the energy infrastructure needed to power them.
I can’t speak to the military applications because, frankly, they were entirely theoretical until recently. Now, the Pentagon is using AI for tasks ranging from intelligence analysis and logistics to battlefield threat identification. There are plans to expand AI’s role in American armed forces. What remains uncertain is how beneficial today’s AI models will ultimately prove to be in warfare – and whether the enormous civilian data-center buildout now underway is necessary to secure that advantage.
The competition with China isn’t imaginary. Both countries are pouring resources into advanced AI, and Washington now openly treats maintaining U.S. leadership as a strategic priority. That doesn’t automatically mean every data center project is justified, but it does explain why federal officials are reluctant to slow the overall buildout.
So far, much of the opposition is rooted in practical local concerns: Electricity demand, utility bills, water consumption, loss of farmland and the effect enormous industrial facilities can have on quality of life nearby. In other words, these aren’t abstract objections to AI. They’re complaints about what happens when a massive new industrial neighbor arrives.
Increasingly, developers are looking beyond major cities for places with large tracts of affordable land and, above all, access to enormous amounts of power. Renewable generation can be attractive, but so can natural gas, nuclear power or simply an existing grid connection with spare capacity. In the AI data center business, access to electricity is rapidly becoming the deciding factor.
As someone who lives in one of these communities, I can tell you the sales pitch isn’t landing very well where I live. It’s a heavily conservative area, but support for Trump doesn’t automatically translate into enthusiasm for a giant server farm moving in next door.
And apparently my neighbors aren’t unusual. Opposition to new data centers now crosses party lines and extends through rural, suburban and urban communities.
Maybe if we were adding reliable power generation – specifically nuclear plants – as quickly as we’re adding electricity demand, people would be somewhat less inclined to complain. Tech companies themselves seem to recognize the problem. Several have already signed long-term nuclear-power agreements to help supply their data centers.
AI centers can create thousands of jobs while they’re being built. The permanent workforce is much smaller. Reuters recently reported that an individual Meta data center typically employs around 100 people long-term, although the number varies based on the size and type of facility.
Now, those can be good jobs – technicians, electricians, engineers, HVAC specialists and security workers, for example. But for a project consuming hundreds of megawatts of electricity and billions of dollars in capital, the permanent head count looks surprisingly small.
There’s user access for the country at large. Many people use Gemini or Grok to do nothing more than making memes. Of course, AI can do far more. Businesses are already using it for software, research, customer service and countless other tasks. However, Deloitte tells us the benefits for businesses have been slow to materialize and hard to measure.
My question is different: How much of that economic benefit flows back to the community hosting the infrastructure?
Datacenter costs and benefits
Tax revenue is probably the strongest argument data-center developers can offer local communities, and in some places it has been substantial. Construction work and permanent jobs matter, too. The harder question is whether those benefits compensate residents for the added demands on electricity, water, land and infrastructure – especially when tax incentives are part of the deal.
I’m not completely dismissing the people who want to promote AI development, but none of them are conjuring up any selling points practical enough or exciting enough to set imaginations aflame. It’s not like the space race of the cold war era.
No one is clutching their pearls in fear of “losing an AI race” that may or may not matter. Promises of an AI-driven productivity revolution are much harder for a town to weigh than the immediate realities of a new industrial facility, higher electricity demand and hundreds of acres of development. Future national benefits may be enormous. But local residents are understandably asking what they receive today.
When I look at the AI debate, I see multiple factions, interests and visions colliding. Plenty of people promoting data centers sincerely believe AI will produce enormous economic and technological benefits. Plenty of people opposing them aren’t anti-technology – they’re homeowners, farmers and ratepayers with legitimate questions about land, water, electricity and costs.
That distinction matters. This debate gets a lot easier to understand once we stop assuming everyone on the other side has sinister motives.
There is another problem, though: The incentives aren’t evenly distributed. Technology companies have enormous sums invested in accelerating AI development. Federal officials see a strategic competition they don’t want America to lose. State and local governments see potential tax revenue.
The people living next door to these facilities have a different set of incentives. They’re thinking about their utility bill, their water supply, their property and the character of their community. None of those groups has to be evil or stupid for their interests to collide.
It’s hard to say where this goes from here, but public opinion has moved sharply against local data center construction. An NBC News Decision Desk poll released in September found 69% of Americans opposed an AI data center being built in their local area. A separate University of Pennsylvania survey found opposition rising from 49% early this year to 61% by summer.
Public opinion can change, of course. But trying to force these projects into unwilling communities without addressing their concerns about power, water, land and local costs seems far more likely to deepen the backlash than solve it.
And that’s really the larger point.
Maybe the AI buildout delivers everything its advocates promise. Maybe the benefits eventually justify the enormous investments in land, power and infrastructure. Or maybe we discover (exactly as we have during previous technology booms) that enthusiastic promises got way ahead of reality.
I don’t know which way this goes. What I do know is that ordinary Americans have very little control over the outcome – even though they’ll ultimately share in the costs, whether or not they also share in the benefits.
That’s one reason I believe diversification matters. When so much of the economy is being reshaped around expensive technologies with uncertain payoffs, it makes sense to consider keeping a portion of your savings in something tangible that doesn’t depend on an AI forecast, a data center or somebody else’s business model.
The dollar buys less today than it did a year ago, a decade ago, a generation ago. That slide is the predictable result of endless printing and borrowing. Dr. Ron Paul believes the FED has the answer. But they’re keeping it to themselves. Fortunately, a new financial chart has publicly exposed their secret moves. See it for yourself: https://freekit.birchgold.com/lf/ro...D_v02a_article&placement=article&cid=rp_media
To learn more about how physical gold could help protect your retirement portfolio, click here to get your FREE info kit on Gold IRAs from Birch Gold Group. And now introducing a Crypto IRA to capitalize on the fastest growing market in the world.
Tweet
Continue reading...
[ H/T The Burning Platform ]
