I live in a town in the Midwest that just voted down a data center project.
Personally I think it's mostly a proxy vote against bigtech/social-media. People are pretty fed up with their practices but don't have power to act at a national level. But, they DO have power at the local level to show up to town council and talk directly (in-person) to their representatives.
I think the other side of this is that there's this old idea (mostly correct) that municipalities partnering with businesses is good for the community because it brings positive side-effects: jobs, more cashflow in the local economy, etc. This is much less true for data centers. It's just a building that uses power and produces heat/by-products. Generally, employment gains are tiny compared with the old "automaker" labor model of the 1960s-1980s
People recognize this and they're not happy. They don't think it's a good deal for their communities.
One argument I've heard that people try to use as proponents of these datacenters is that they bring in employment in construction and electrical engineering even if it's temporary. Their argument falls apart very quickly when it's pointed out that because of the specialization many owners of these locations hire specific firms and contractors willing to travel across the nation for the work instead of hiring a local construction company or an on-site engineer. Some locations are managed entirely remotely where one engineer handles multiple sites, and the only people actually there are security, maintenance, and cleaning staff who might also be traveling contractors that move between locations in a circuit. That's like six people for something with a footprint the size of fifty family homes.
Even if it were true, I don't understand how this argument makes sense: So assume the construction were entirely done by local businesses/hires. That gets a short-term employment boost while the facility is being built. And then what? The construction is finished, the employment dries up again, but the data center is still there, with all the downsides.
Also, you could get the exact same short-term boost by building something else, e.g. housing.
Offsetting long-term costs with short-term benefits doesn't seem like a good strategy.
If your local governments are funded by property taxes, Data centers can bring in new revenue while requiring few services. Which could be used to shore up stressed public budgets or fund other economic development activities to bring in jobs.
These projects can be developed and located responsibly, and every project is different. I don't think a blanket ban is good policy.
This is a major reason I think SpaceX’s space DCs aren’t insane, this is a pretty clear trend. These things don’t bring durable employment, they concentrate costs on infrastructure locally, and their benefits go to the world, so the only way I’ve seen the calculus work for a local community are if they levy property tax on the contents of the DC, use it to subsidize your local property taxes/infrastructure, and then foist the cost due to increased power demand on the wider region. My understanding is this is what Loudon County, VA does with its many DCs, taking the benefits and spreading the costs across the entire PJM region. You effectively have poor Baltimoreans subsidizing the highest median income county in the US via increases in their heating bills. Of course, the rest of the PJM region is annoyed about this and starting to try to obstruct that.
The world is large. There is still infinite room left on earth closer than space.
The reason the DC want to be placed in proximity to people is to get access to the infrastructure, electricity grid and roads. I think the next natural step is to just put the DC further away from existing infrastructure and pay for the connect, not move to space.
The concern isn't at all about physical space, DCs are compact, it's about limitations of the infrastructure you're talking about. Turbines currently have a multi-year backlog. The interconnection queue for new generation is an even longer backlog in many grid regions. And there's a general populist backlash against them.
I've been interested in understanding what would make people more amenable to data centers. We kind of need them, though arguably many of the ones being built now are motivated by foolish AI bubble incentives.
Quieter? Lower water use? Lower energy use? Mandatory accessory green spaces? Property taxes that reflect the value being derived relative to inconvenience/pain inflicted on the community? Jobs programs?
I think there's a lot of ideas to mitigate the downsides of data centers. Many of the people who don't want data centers have such proposals that are opposed by different people who don't want data centers.
It would help a ton if they'd pull some strings to get more power added to the grid than they consume so they don't cost residents and local businesses money. Or they could pay to subsidize residential usage and keep rates low.
They could give all nearby residents free/subsidized solar + battery for their homes.
This empowers people making them feel less beholden to rising energy prices, and gives the data center more energy for its needs as the grid is freed up.
> I've been interested in understanding what would make people more amenable to data centers.
We've been slowly boiling alive in the reality that the tech industry has long been evolving to hurt us more and help us less each year. We'd be neutral or welcoming to data centers if we didn't know that storing and processing all that data was going to be used against us.
There's no good reason to build datacenters in many of the places people are trying to build them. Look at all the ones they're trying to build in Phoenix or Tucson, Arizona, or in New Mexico. These are deserts! They're the exact opposite of a place where it makes sense to build a datacenter.
It's not enough to offer incentives. You have to explain why you even need to do it in the first place, and the answer better not be "we want money."
These are good places to build a data center if you're planning on operating it entirely off-grid with a photovoltaic solar panel array (or even just augmenting grid power with PV solar generated onsite). Your cooling costs increase somewhat compared to a cooler outdoor climate, but the desert southwest is basically the best place in the US to put solar panels.
Off grid doesn't make sense (computers ramp up and down extremely quickly which makes balancing an islanded DC challenging), DCs function a lot better if they can utilize the inertia of the grid.
Many projects are bringing their own generation, but still trying to get an interconnection.
> We kind of need them, though arguably many of the ones being built now are motivated by foolish AI bubble incentives.
I think the purpose is important. People understand the need for power stations or garbage dumps or waste treatment plants. However:
- Those are clearly understood to be a negative for the immediate environment. No one wants to live next to a garbage dump.
- They are just as clearly understood to be a benefit or even necessity for overall society, including the people having to live close by: Everyone wants pickup and processing of their garbage working smoothly.
With AI datacenters, the situation is weirdly turned on its head:
- The overall necessity of AI is still deeply questionable. Businesses and governments are pushing it and yes, there are a lot of users, but overall sentiment of the population seems to be ambivalent to strongly negative. There is everything from warnings of addiction and AI psychosis to AI simply destroying most peoples' livelihood. So there is no "greater cause" for which someone should tolerate an AI datacenter in their neighborhood, the way they might tolerate a wind turbine.
- To make up for this, AI proponents seem to try and conjure up some ostensible direct benefits the data center (and not the AI wave in general) would have for the neighborhood, like jobs or followup businesses. It increasingly gets clear that this is just a ruse, maybe an exploitation of people who confuse a data center with a research campus.
It's like trying to sell a fentanyl factory by saying its exhausts contain CO2 and so might stimulate local plant life.
So my feel is, data centers will always have some negative effects on the neighborhood. But how far the neighborhood is willing to tolerate those effects is directly dependant on its view of AI in general.
To start, there is a lot of misinformation out there and the NDAs that surround data center construction and operation don't help. People will cite water consumption as a huge problem when modern hyperscalers use substantially less water because they're now using closed loop cooling instead of evaporative cooling. You'll see people cite noise because they saw a video online of a crypto mining grifter who bought a bunch of shipping containers and haphazardly threw together air cooled mining rigs with 80mm fans screaming away. I even saw one video of a woman who claimed data centers gave her diabetes despite the fact that she was obese.
Amazon and other companies already have job training programs because they cannot find enough skilled labor to build and operate their data centers. The number of jobs commonly cited are comically lower than what is common to operate a modern hyperscaler. In my experience, hyperscalers often have at least 100-200 people on site to operate the data center and I've seen more than 1000 people on a site when the data center is under construction.
The real issue, as always, are the local governments and utilities that sellout out the citizens and fail to create and enforce building codes. The governments should be using the demand for data centers to partner with the companies and have them pay to modernize and fix the power grid. They should be using them to help subsidize green energy initiatives among other things and fund other projects to benefit the community.
The inconvenient truth is that the problem with data centers lies with the people in the communities who continue to elect politicians who, time and time again, make decisions counter to the best interests of their community. Data centers just happen to be the latest scapegoat to distract people from corrupt politicians and an community that is not civically engaged enough to hold their politicians accountable.
> the people in the communities who continue to elect politicians who, time and time again, make decisions counter to the best interests of their community.
I wonder where all the politicians are who would make decisions for their communities instead...
The core issue is that most Americans aren't civically engaged. They don't take the effort to track what's happening within their government which then means that they don't know and are unable to hold their elected officials accountable.
Communities can have more effective politicians, but that means voting out individuals who make bad decisions or decisions that are counter the will good of the community they serve.
I'm pretty sure it's mostly ideological problems with AI, rather than any concrete impact on the communities data centers are near. Data centers are just not that big or impactful a structure, and they're also not particularly new. What is new is effective AI being a force visible to the lives of people, including activists of various stripes who who are threatened by it and want to work against it by attacking data centers.
Honestly, I think it's possible that there isn't really organic opposition to data centers from people living in communities near them at all - but instead, there is opposition to them from people like Mother Jones magazine journalist Sophie Hurwitz, who is therefore motivated to write an article reporting on Maine banning data centers while framing it as a reasonable policy, and implying that this is a reasonable thing for other local governments considering data center bans to enact.
I note that every person cited in this article is some kind of national-level ideological actor - a member of a pro-data-center lobbying group; a researcher affiliated with the Federation of American Scientists which is a NGO headquartered in Washington D.C.; the head of Good Jobs First, which is another Washington D.C.-headquartered nonprofit; and several well-known national politicians who are already known for being suspicious of the tech industry.
There's no quote from any ordinary person in Maine who talks about some concrete negative impact of a data center near them - the closest thing is a link to an article with a quote from the Maine state representative who sponsored the bill, which states: "“It’s not that there’s no place for data centers in Maine,” said Democratic Rep. Melanie Sachs, who sponsored the measure. “Frankly, the tradeoffs have not been shown to be of benefit to our ratepayers, water usage or community benefit in terms of economic activity.”. The idea that data centers use a particularly large amount of water is basically complete bullshit promulgated by national-level prestige journalists (https://www.andymasley.com/writing/the-ai-water-issue-is-fak...), which makes me skeptical that this is an issue brought to Sachs' attention from her local constituents.
Data centers really do use a lot of electrical power, but the article has the quote "n Maine, electricity bills have already increased by 58 percent on average over the last 5 years. Much of that price jump is likely due to the state’s reliance on natural gas—but some Mainers fear that data center buildout will only increase their expenses.", which is the sort of thing you'd write if you were trying to associate a rise in electricity costs with data centers without being able to demonstrate that data centers are actually causing electric power costs to increase for ordinary people.
I don't understand why these things need to be built in populated areas.
If it's true that they can be constructed in space and operated remotely, then they can also be placed on container ships, on isolated ocean platforms like oil rigs, or in unpopulated areas on land.
If it's not true that they can be constructed in space, then we'd probably better stop telling ourselves that it's possible.
I live in a town in the Midwest that just voted down a data center project.
Personally I think it's mostly a proxy vote against bigtech/social-media. People are pretty fed up with their practices but don't have power to act at a national level. But, they DO have power at the local level to show up to town council and talk directly (in-person) to their representatives.
I think the other side of this is that there's this old idea (mostly correct) that municipalities partnering with businesses is good for the community because it brings positive side-effects: jobs, more cashflow in the local economy, etc. This is much less true for data centers. It's just a building that uses power and produces heat/by-products. Generally, employment gains are tiny compared with the old "automaker" labor model of the 1960s-1980s
People recognize this and they're not happy. They don't think it's a good deal for their communities.
One argument I've heard that people try to use as proponents of these datacenters is that they bring in employment in construction and electrical engineering even if it's temporary. Their argument falls apart very quickly when it's pointed out that because of the specialization many owners of these locations hire specific firms and contractors willing to travel across the nation for the work instead of hiring a local construction company or an on-site engineer. Some locations are managed entirely remotely where one engineer handles multiple sites, and the only people actually there are security, maintenance, and cleaning staff who might also be traveling contractors that move between locations in a circuit. That's like six people for something with a footprint the size of fifty family homes.
Even if it were true, I don't understand how this argument makes sense: So assume the construction were entirely done by local businesses/hires. That gets a short-term employment boost while the facility is being built. And then what? The construction is finished, the employment dries up again, but the data center is still there, with all the downsides.
Also, you could get the exact same short-term boost by building something else, e.g. housing.
Offsetting long-term costs with short-term benefits doesn't seem like a good strategy.
This is a good take.
Put a different way, some companies have made a lot of money with business models that hinge on victims never being able to reach a human.
Those same companies want to set up phone centers in the neighborhoods of the people they’ve neglected that also will not take their calls.
Town hall it is.
If your local governments are funded by property taxes, Data centers can bring in new revenue while requiring few services. Which could be used to shore up stressed public budgets or fund other economic development activities to bring in jobs.
These projects can be developed and located responsibly, and every project is different. I don't think a blanket ban is good policy.
This is a major reason I think SpaceX’s space DCs aren’t insane, this is a pretty clear trend. These things don’t bring durable employment, they concentrate costs on infrastructure locally, and their benefits go to the world, so the only way I’ve seen the calculus work for a local community are if they levy property tax on the contents of the DC, use it to subsidize your local property taxes/infrastructure, and then foist the cost due to increased power demand on the wider region. My understanding is this is what Loudon County, VA does with its many DCs, taking the benefits and spreading the costs across the entire PJM region. You effectively have poor Baltimoreans subsidizing the highest median income county in the US via increases in their heating bills. Of course, the rest of the PJM region is annoyed about this and starting to try to obstruct that.
The world is large. There is still infinite room left on earth closer than space.
The reason the DC want to be placed in proximity to people is to get access to the infrastructure, electricity grid and roads. I think the next natural step is to just put the DC further away from existing infrastructure and pay for the connect, not move to space.
The concern isn't at all about physical space, DCs are compact, it's about limitations of the infrastructure you're talking about. Turbines currently have a multi-year backlog. The interconnection queue for new generation is an even longer backlog in many grid regions. And there's a general populist backlash against them.
I've been interested in understanding what would make people more amenable to data centers. We kind of need them, though arguably many of the ones being built now are motivated by foolish AI bubble incentives.
Quieter? Lower water use? Lower energy use? Mandatory accessory green spaces? Property taxes that reflect the value being derived relative to inconvenience/pain inflicted on the community? Jobs programs?
I think there's a lot of ideas to mitigate the downsides of data centers. Many of the people who don't want data centers have such proposals that are opposed by different people who don't want data centers.
It would help a ton if they'd pull some strings to get more power added to the grid than they consume so they don't cost residents and local businesses money. Or they could pay to subsidize residential usage and keep rates low.
They could give all nearby residents free/subsidized solar + battery for their homes.
This empowers people making them feel less beholden to rising energy prices, and gives the data center more energy for its needs as the grid is freed up.
> I've been interested in understanding what would make people more amenable to data centers.
We've been slowly boiling alive in the reality that the tech industry has long been evolving to hurt us more and help us less each year. We'd be neutral or welcoming to data centers if we didn't know that storing and processing all that data was going to be used against us.
There's no good reason to build datacenters in many of the places people are trying to build them. Look at all the ones they're trying to build in Phoenix or Tucson, Arizona, or in New Mexico. These are deserts! They're the exact opposite of a place where it makes sense to build a datacenter.
It's not enough to offer incentives. You have to explain why you even need to do it in the first place, and the answer better not be "we want money."
These are good places to build a data center if you're planning on operating it entirely off-grid with a photovoltaic solar panel array (or even just augmenting grid power with PV solar generated onsite). Your cooling costs increase somewhat compared to a cooler outdoor climate, but the desert southwest is basically the best place in the US to put solar panels.
Off grid doesn't make sense (computers ramp up and down extremely quickly which makes balancing an islanded DC challenging), DCs function a lot better if they can utilize the inertia of the grid.
Many projects are bringing their own generation, but still trying to get an interconnection.
> We kind of need them, though arguably many of the ones being built now are motivated by foolish AI bubble incentives.
I think the purpose is important. People understand the need for power stations or garbage dumps or waste treatment plants. However:
- Those are clearly understood to be a negative for the immediate environment. No one wants to live next to a garbage dump.
- They are just as clearly understood to be a benefit or even necessity for overall society, including the people having to live close by: Everyone wants pickup and processing of their garbage working smoothly.
With AI datacenters, the situation is weirdly turned on its head:
- The overall necessity of AI is still deeply questionable. Businesses and governments are pushing it and yes, there are a lot of users, but overall sentiment of the population seems to be ambivalent to strongly negative. There is everything from warnings of addiction and AI psychosis to AI simply destroying most peoples' livelihood. So there is no "greater cause" for which someone should tolerate an AI datacenter in their neighborhood, the way they might tolerate a wind turbine.
- To make up for this, AI proponents seem to try and conjure up some ostensible direct benefits the data center (and not the AI wave in general) would have for the neighborhood, like jobs or followup businesses. It increasingly gets clear that this is just a ruse, maybe an exploitation of people who confuse a data center with a research campus.
It's like trying to sell a fentanyl factory by saying its exhausts contain CO2 and so might stimulate local plant life.
So my feel is, data centers will always have some negative effects on the neighborhood. But how far the neighborhood is willing to tolerate those effects is directly dependant on its view of AI in general.
To start, there is a lot of misinformation out there and the NDAs that surround data center construction and operation don't help. People will cite water consumption as a huge problem when modern hyperscalers use substantially less water because they're now using closed loop cooling instead of evaporative cooling. You'll see people cite noise because they saw a video online of a crypto mining grifter who bought a bunch of shipping containers and haphazardly threw together air cooled mining rigs with 80mm fans screaming away. I even saw one video of a woman who claimed data centers gave her diabetes despite the fact that she was obese.
Amazon and other companies already have job training programs because they cannot find enough skilled labor to build and operate their data centers. The number of jobs commonly cited are comically lower than what is common to operate a modern hyperscaler. In my experience, hyperscalers often have at least 100-200 people on site to operate the data center and I've seen more than 1000 people on a site when the data center is under construction.
The real issue, as always, are the local governments and utilities that sellout out the citizens and fail to create and enforce building codes. The governments should be using the demand for data centers to partner with the companies and have them pay to modernize and fix the power grid. They should be using them to help subsidize green energy initiatives among other things and fund other projects to benefit the community.
The inconvenient truth is that the problem with data centers lies with the people in the communities who continue to elect politicians who, time and time again, make decisions counter to the best interests of their community. Data centers just happen to be the latest scapegoat to distract people from corrupt politicians and an community that is not civically engaged enough to hold their politicians accountable.
> the people in the communities who continue to elect politicians who, time and time again, make decisions counter to the best interests of their community.
I wonder where all the politicians are who would make decisions for their communities instead...
The core issue is that most Americans aren't civically engaged. They don't take the effort to track what's happening within their government which then means that they don't know and are unable to hold their elected officials accountable.
Communities can have more effective politicians, but that means voting out individuals who make bad decisions or decisions that are counter the will good of the community they serve.
Data centers are fundamentally a net negative to the region, value extraction towards the oligarchs.
Why would anyone want them?
The only thing that'd change my mind would be full communal ownership in addition to everything you've said.
I'm pretty sure it's mostly ideological problems with AI, rather than any concrete impact on the communities data centers are near. Data centers are just not that big or impactful a structure, and they're also not particularly new. What is new is effective AI being a force visible to the lives of people, including activists of various stripes who who are threatened by it and want to work against it by attacking data centers.
Honestly, I think it's possible that there isn't really organic opposition to data centers from people living in communities near them at all - but instead, there is opposition to them from people like Mother Jones magazine journalist Sophie Hurwitz, who is therefore motivated to write an article reporting on Maine banning data centers while framing it as a reasonable policy, and implying that this is a reasonable thing for other local governments considering data center bans to enact.
I note that every person cited in this article is some kind of national-level ideological actor - a member of a pro-data-center lobbying group; a researcher affiliated with the Federation of American Scientists which is a NGO headquartered in Washington D.C.; the head of Good Jobs First, which is another Washington D.C.-headquartered nonprofit; and several well-known national politicians who are already known for being suspicious of the tech industry.
There's no quote from any ordinary person in Maine who talks about some concrete negative impact of a data center near them - the closest thing is a link to an article with a quote from the Maine state representative who sponsored the bill, which states: "“It’s not that there’s no place for data centers in Maine,” said Democratic Rep. Melanie Sachs, who sponsored the measure. “Frankly, the tradeoffs have not been shown to be of benefit to our ratepayers, water usage or community benefit in terms of economic activity.”. The idea that data centers use a particularly large amount of water is basically complete bullshit promulgated by national-level prestige journalists (https://www.andymasley.com/writing/the-ai-water-issue-is-fak...), which makes me skeptical that this is an issue brought to Sachs' attention from her local constituents.
Data centers really do use a lot of electrical power, but the article has the quote "n Maine, electricity bills have already increased by 58 percent on average over the last 5 years. Much of that price jump is likely due to the state’s reliance on natural gas—but some Mainers fear that data center buildout will only increase their expenses.", which is the sort of thing you'd write if you were trying to associate a rise in electricity costs with data centers without being able to demonstrate that data centers are actually causing electric power costs to increase for ordinary people.
Mainframe centralized computing like the good old days in a way that will make OpenAI profitable is not coming back, IBM and Digital are gone.
I don't understand why these things need to be built in populated areas.
If it's true that they can be constructed in space and operated remotely, then they can also be placed on container ships, on isolated ocean platforms like oil rigs, or in unpopulated areas on land.
If it's not true that they can be constructed in space, then we'd probably better stop telling ourselves that it's possible.
Related:
Maine is about to become the first state to ban major new data centers
https://news.ycombinator.com/item?id=47708817
First data centers, next up: maineframes
[flagged]