I cannot figure out if this data center panic is legit or trumped up. There are legitimate nation-state and political entities that would love to create pressure for degrowth in this area. And, pollution and natural resources are a concern, yes, fine, sure. NIMBY happens, I get it ...
*BUT*, we've been building huge data centers for decades and they are minuscule compared to the size of the earth. My skeptic meter maxed out when i saw a Tesla with a "End Data Centers" bumper sticker on it. This is just the latest popular panic, isn't it?
We’re building data centers in the U.S. at an unprecedented and increasing rate [1]. Is it so hard to imagine that something which was fine at one scale may begin to cause issues at some massively higher scale?
If I drink one glass of wine per day, it’s probably fine. If I drink three bottles per day, I’m gonna start having problems.
Yeah, but when we account for that scale, and look at resource usage, datacenters aren't topping the charts, or even really getting in the top 5, for anything.
You have people who eat meat at every meal losing their mind over water usage and people who set their AC to 68 panicking over energy usage.
Eating meat fills my belly. Setting my AC to 68 keeps me cool. A data center costs my community money, damages the environment directly and indirectly, and further enriches a few already rich people.
AC at 68F is obnoxious and wasteful. At least data centres enable me to communicate with distant colleagues, watch enriching movies, and use very strong AI tools to accelerate my research.
Complete tangent, but my impression is that one glass of wine per day is "fine", but you're still statistically better off in a non-trivial way by going down to zero. Not counting the benefits of course.... :)
To continue to take us off topic, there is actually pretty limited evidence that this is true.
It is true that wine doesn't do anyone any good (unless you count it making people more social or something). But the effects of one glass a night is indistinguishable from zero. There are definitely statistically significant morbidity effects at 3.
I've been shocked reading public health proclamations that "anything above zero" is bad, and then reading the actual underlying paper they refer to, which clearly states exactly the opposite finding. I think it is an extension of the public policy belief that the public cannot be trusted with moderation, and so therefore must be communicated to in absolutes, and nobody will read the underlying paper anyway.
Some of it's a panic, but I'm leaning towards many of the concerns being true or at least partially true due to the sheer scale of the construction. The figures I've been seeing are something like half of all current private construction projects are data center related these days and it's the biggest single sector of construction in terms of money spent.
It's also a lot easier to protest data centers then AI companies and they generally provide minimal benefits to the local communities (a brief job boom for construction but minimal reoccurring benefits afterwards) and quite a few negatives like power consumption and noise. People can't stop their jobs from replacing them with an LLM, but they can at least stop the local data center from being built and raising the electricity bill by $70 a month[0]. This batch of data center construction is essentially yet another physical representation of the (often negative) changes the general populace faces with AI.
[0] $70/month is how much my old power company said they'd have to raise prices if several proposed data centers were built. I was already paying $150-200 a month to power a 1 bed apartment so you can bet your ass I filed a complaint with the county about it.
> they generally provide minimal benefits to the local communities
Data centers pay Loudoun County, Virginia $26 in taxes for every $1 in services they consume — nearly $1.3 billion, 45% of the county's budget — which is why Loudoun homeowners pay the lowest property tax rate in Northern Virginia.
What I don't understand is why communities want to block the data centers.
Just make them pay!
Your electricity price shouldn't go up because of them. So just make a rate card for data centers that ensures the burden doesn't fall on residents. That's what Oregon started doing, and residents' electricity prices actually started falling, because the fixed costs of energy production were being spread across more kilowatt-hours and the data centers weren't getting a special deal.
I've also seen counties offer tax incentives. Why? Just have them pay property tax. Then you get nicer schools, more police, better roads.
Instead we have these campaigns to stop them, instead of making them a local cash cow. Loudon County gets half of all its revenue from data center property tax. Its residents were already rich, and now their local taxes are artificially low. Who wouldn't want that?
Wait… large customers move in and your rate goes up?
Forget about it. These AI companies could really help themselves by subsidizing local rates. Imagine if a large AI datacenter moved in AND your rate dropped by $70!
Oregon just passed a law saying data centers will pay more for electricity than normal people, and the math says that everyone's rates will be going down.
So, again, I think it depends on whether you live in a red or blue state.
*As high as 2.2 °C, with average downwind air temperatures 0.7–0.9 °C warmer than corresponding upwind areas. Thermal signatures were detectable at distances up to 500 m"
"As high as", combined with "up to" show two upper limits, but you can be sure if they were simultaneous they would have said so, especially when considering their prhasing of downwind being warmer comes after mentioning the upper limit. It really seems like you are supposed to overlook that it's an upper limit and subconsciously add another 0.7-0.9 on top.
What they are saying is you can feel the warm breeze coming off the hot thing if you stand near enough.
A comparison to other industrial sites would put it in perspective. As would a distinction between localised warming which is what is happening here compared to global warming where small temperature differences at a global scale are far more significant. The publicity about global warming risk has put a lot of effort attuning people to the significance of those small temperature shifts, to co-opt that education to make people fear a situation where it doesn't apply is undermining their work. That puts us at more risk than any one data centre. People can observe a local increase of 2 degrees in relative comfort and think that it won't be catastrophic at a global scLe.
Seems to me that the problem isn't that the data centres are being built. It's that the people building them aren't paying for their full cost. Despite the organisations building them being some of the richest in the world that can absolutely afford to do so.
(externalities in terms of environmental impact and resource usage (water, electricity, etc) usage driving up costs for everyone else).
I'm a hobbyist IT enthusiast, not a professional computer guy, but from my point of view there's a big difference between a bunch of servers hosting something people want, like Netflix, and a warehouse full of GPUs running around the clock to make Lobster Jesus or whatever.
It's also a proxy for people's feelings on AI--essentially the only way people have to fight back against the new technology that makes eternal garbage and is supposedly coming for their jobs is to push for banning "data centers."
Totally agree that it's just a proxy for people's feelings on AI, but my Lobster Jesus is much more important to me than a Netflix show! All of the datacenter usage (for inference) is just serving something people want. Time has already shown that the datacenter for training is also just creating something people want.
How is there a big difference between those two things? People want Netflix, people also want the output of AI whether that is lobster Jesus or something else. This seems more like substituting what people want, which clearly includes AI, for what some people think others should want. I can guarantee you there are people who don’t want Netflix and think it’s a waste of resources that encourage people to binge TV, become a couch potato, and end up being drags on the healthcare system. Should their preferences override those who do want it?
Democracy, even at its most ideal, is often about overriding preferences. Netflix helps to float an entire industry of creators and a tangible pop culture. It's not entirely clear (or softened, maybe not clear to communities) what AI build out floats currently except AI companies and the tangible negatives of generative AI.
I think that potentially the difference is that, while people do in some sense want AI, they don't want it in the same way that they want Netflix. I'd compare it to Amazon: it perfectly possible for people to like the ability to like 2 day shipping while also liking that the site is flooded in cheap crap.
So maybe it is that people like AI in certain domains but HATE the way that it is being used overall?
> How is there a big difference between those two things?
Netflix just serves existing data so their biggest need is data storage ala CDN for closer retrieval to the user. Very low requirements. Creating sweet baby lobster jesus requires incredible amounts of computer/power/cooling. The requirements are totally different. Yes, both ultimately deliver pixels to consumer's eyeballs, but how they get there are so different that it's laughable that people cannot see the difference.
There's also different building pressures. Before, companies would want to build / power / maintain the smallest amount of data center hardware they could while still serving their needs. Overprovisioning just meant worse opex number.
Now, because supply is so constrained relative to demand, there's no limit. More is always better, which leads to a crazy amount of construction (as its negative effects).
> there's a big difference between a bunch of servers hosting something people want, like Netflix, and a warehouse full of GPUs running around the clock to make Lobster Jesus or whatever.
ChatGPT has been the most downloaded app every month for a year and a half.
The conflating of a data center and an AI/GPU farm is disingenuous at best, and the AI people know what they are doing. None of the "traditional" data centers are being powered on site with gas turbine generators. New GPU farms are also being designed with much larger footprints with much more needs from the public utilities.
AI inference and training GPUs require way, way more power than anything manufactured before. Like, between 2020 and 2026 we went from ~400W for an A100 to ~1400W for an NVIDIA B300. And you cannot make that happen without changing the datacenter itself. More noise, way more complicated cooling setup, new power sources, etc. Not all datacenters are created equal.
The financing for datacenters is also very shady and its own can of worms
I don't think it's necessarily shady, but let's say that we're assisting at bets and capex spending we've never seen in history.
Companies like Meta, Alphabet, etc that were drowning in cash flow and buying back their stock are now back at issuing shares, creating a huge number of SPVs, issuing debt, etc.
There's a gigantic spending spirale with no signs of ending. Nvidia is financing companies to buy its own hardware.
Mind you, this is nothing new in some businesses. Airbus and Boeing have been lending money to companies to buy their own airplanes forever.
But its just all too much, with very little money coming. To understand how massive is the spending: even if every adult on the planet spent 250$ on AI per year this would still not cover anywhere near what's being invested.
Just because we've been doing something forever does not mean that a recent resurgence in policy interest and a re-evaulation of the social worth of these projects is a waste of time.
Part of the problem is AI datacenters make bad neighbors.
Because they are in such a rush to become operational they cut corners.
Instead of AC they use swamp coolers draining aquifers and thus wells of the neighbors.
Instead of using new power plants (AKA using renewables as most new plants do, and all old datacenters did). They either hook up to the grid and jack up everyones electric prices or use generators because they can't be bothered to wait for a grid connection.
All of which is to say while on the grand scale they may be fine. Locally they can be unpleasant to have in your back yard.
If the LLM datacenter builders and operators weren't cutting corners and rushing and doing everything they can to be good neighbors. There would be much less push back.
I say this as someone with 2 huge pre-AI datacenters in my town that have been fine.
It's just the outrage story of the month. There will be some new moral panic in September and we'll all move on. The attention span for topics like this on social media is quite short.
That's where the resources are. You wouldn't expect some uber wealthy billion dollar company to have to provide power/water needs way out in the middle of nowhere would you?
One strange form that the post-truth era takes is deciding that stances you don't believe in aren't even real. Saves you the trouble of having to engage with them meaningfully.
These days there always seems to be panic regardless. I find it better to look at the merits of individual claims rather than try to group everything together into a binary answer like a possible conspiracy on one thing means nothing else needs to be considered.
Zoning, grid strain (->pricing), and chip strain (->pricing) seem to be pretty legit most of the time I look into them. Noise seems to have a lot of "they opened a 5G tower in the town and now I get nightmares" type panic but also some legitimate problems which, ultimately, relate to zoning. I try to reserve NIMBY for "things I want in someone's back yard, just not mine". The reality is data centers don't fit that because there is no reason they need to be in anyone's back yard. They are perfectly happy to be built in the middle of other industry/commercial blocks or even in the middle of nowhere.
I think if the grid had been better prepared for massive demand growth with cleaner sources than before and things came in slow enough that zoning didn't feel rushed people wouldn't really make as much of a stink around it. Unfortunately, the US spent many years with flat grid demand and companies want to cram these DCs in as fast as possible because their demand was for last year already.
For some large factor, it's also a reason to really hate on AI. Which there may well be valid reasons for, I just also wonder how many of these people actually care about the environment and how many are just saying it because it's another con to list.
My great-granddad was unconcerned about the deforestation of the Amazon during the 1970s until the little forest on the way out of his small town was cut down for a new subdivision.
One HUGE issue is that datacenters don't have an afterlife in the communities where they're located.
They add to the Rust Belt once their lifetimes end, contrast this with old brick warehouses which were eventually repurposed as homes and offices.
If they were built with a bit of thought around what happens once they depreciate completely and they're abandoned/sold for residual value they wouldn't be as despised.
Current construction completely ignores the nearby communities present and future development.
If the ai bubble bursts, these buildings will be able to be repurposed unlike rust belt factories. With all of that cooling, you could turn them into a weird form of multi-family housing. Or use them for indoor skate pars, kart racing tracks, whatever you can imagine. I don't think rust belt is an appropriate comparison. Also, these places don't hire that many people like a factory did. If the factory shut down, entire towns could collapse with it. If a data center shut down, maybe 20 people would notice????
> Current construction completely ignores the nearby communities present and future development.
The issue is an incentive one, for the companies involved in the construction.
Right now, they are getting paid so much to build it fast. Time is of the greatest importance. Nothing else is mattering to anyone involved except how fast datacenters are being built. At best, this is an after-thought.
As such, I believe that we have to change the incentives to create more sustainability but it requires time, discussion. Perhaps laws can help in making them more sustainable but currently, the sad reality is that for all the parties actually involved, I don't think that many truly care about the after-math or repurcussions of it as much as its the only thing that they want right now
Basically, anything people decide not to like is targeted for its climate impact
Some of that targeting is valid and points out optimizations that can be done
They want you to ignore their air conditioning and fridge’s greater relative impact, because the thing they dont like has “no use” in comparison to the thing they do like. They also believe that any power use is easily transportable to the use they do like, which is usually false.
But there are configurations of data centers with less environmental impact, which should become ubiquitous
The outrage is real to the individual, but manufactured. Media (e.g. 404media, Ed Zitron) promotes anything that engages for clicks, views, and subscriptions. Other parties do it to hinder competition, weaken the US, or distract from much worse polluters.
It kind of feels to me like a big element of this is that the social contract in America has broken down. Why should an individual accept any sacrifice whatsoever if all of the profit is going to go to the billionaires who own it? If they do get built and someone suffers an externality, why would they believe that society will do anything to make things right?
Basically, I think that the the capitalist class are finding out that things like social safety nets, socialized healthcare, income equality, etc. aren't just important for humanitarian reasons; they are also prerequisites for people to trust society enough to be willing to coordinate around major projects.
I think that's another example of the social contract breaking down. Lots of people either cannot afford to invest or no longer believe that investing will allow them to afford things like a house or retirement that were traditionally part of being middle class.
I imagine that things would be dramatically different if most people believed that the success of data centers would directly lead to their economic success.
I've done a bit of research into it, and I think it's a mix.
The energy thing is exaggerated. Want to compensate for a week of heavy AI coding usage? If you drive to work, telecommute for one day. Skip one long-distance DoorDash order. Cars use tons of energy, a lot more per-user than the fraction of data center capacity you're using with your AI stuff. A burrito on DoorDash delivered using an average MPG gas car uses something like a week's worth of heavy AI usage in energy.
The water thing is region-specific and where is a problem it's due to incompetent or corrupt goveranance. Bottom line is that if a data center is using too much water, it means they're not paying enough for that water. Make them pay more for their water and they'll magically discover closed cycle cooling.
The noise thing is in some cases real, and needs to be fixed with local regulation or a class action suit by neighbors to force them to deal with the noise problem.
I am a little puzzled though by where these things are being built. Phoenix? Other than cheap solar power I can't tell why you'd want a data center there given the energy needed for cooling.
Why are we not putting them up in the Arctic circle? Build run-of-the-river hydro in northern Canada and put them way up there. Free cooling, especially in winter. AI is not a very latency-sensitive load so an extra 50ms doesn't mean much.
I suspect state incentives are a reason. The giant Amazon data center proposal in Texas with the giant natural gas power plant sounds a lot like "taxpayer funded indirect fossil fuel industry subsidy." Offer some kind of big state incentive to get a huge natural gas customer with long-term locked-in demand.
> Two AI Data Centers Require More Energy Than All Wisconsin Homes Combined: Just two AI data center projects — the Microsoft data center in Mt. Pleasant and Vantage data center in Port Washington — will require a combined 3.9 gigawatts (GW) of electric power, which is enough energy to power 4.3 million Wisconsin homes. There are only 2.8 million housing units in the entire state.
> In fiscal year 2023-24, we supplied more than 20,749 gigawatt-hours (GWh) of power for more than 1.6 million electric service customers, including nearly 6,000 in the Owens Valley.
That's an average of 2.3 GW over the year.
-----
> The water thing is region-specific and where is a problem it's due to incompetent or corrupt goveranance. Bottom line is that if a data center is using too much water, it means they're not paying enough for that water.
> When Google purchased a defunct aluminum smelter in The Dalles, it also acquired the right to pump massive amounts of water from underground — up to 3.88 million gallons a day. That’s almost as much water as the entire city averaged per day in 2024.
> The corporation transferred those water rights to the city in 2021, Bentz noted, allowing The Dalles to pump more water from underground. The city only obtained part, not all, of the former smelter’s water rights, Mays said after OPB’s story published.
It's the rights to pump the water. Google can pump up to 3.88 million gallons per day and the city can't bill them a penny for that. Also, because it's an old claim, it's likely more senior to other water rights.
> Make them pay more for their water and they'll magically discover closed cycle cooling.
A closed cycle cooling system doesn't solve the overall problem of water usage. Yes you have a cycle within the datacenter, but you need to then move that energy somewhere else, out of the system. Eventually that energy needs to find its way in the environment. A close loop system for a datacenter reduces the water use on site, but actually increase electro use. If that electricity comes from a power plant that is cooled by water you just moved the problem upstream
That's true. Making them pay enough for water might mean the data center doesn't even get built there. It gets built somewhere more logical with abundant cheap water.
It's gotta be costs, right? Everyone wants this done fast and at large scale, so they're going for any sites with existing large electrical connections I assume.
It is real but it's about the unaccountability, dishonesty, and misanthropy of corporate overlords, rather than about the externalities of data centers per se. See https://news.ycombinator.com/item?id=49180901
It was already a huge problem that powerful people in this world work to enrich themselves with no regard to regular people's wellbeing. The rage about data centers is because it suddenly started happening in everyone's backyard, making the abstract problem into a physically palpable one.
Imo their anger is completely justified. This only gets any better when there is an actual imperical shift towards behaving prosocially and away from mindless profiteering, and people have every right to want that from people in power. I fully expect the path we're on to end in some kind of revolution if nothing changes and there seems to be no sign of anything changing.
In particular, shame on anyone who uses arguments of the form: "it's these company's jobs to maximize profit, therefore they should not be beholden to behaving prosocially; the abstract force of economic growth justifies all iniquities because it will miraculously benefit everyone eventually". This argument is a lie on every count: it's only their job because we've chosen for it to be their job; the force of economic growth famously hurts people all the time; there is no evidence that it will fix everything versus just accelerate burning everything down (quite literally in this case); and finally, people's happiness is proportional more to their ability to trust those in power to care about them and their communities than to their material wealth per se, so trust is the thing they want maximized, not profit--and they are presently starved of it, because the people with power behave like lying manipulative scumbags at every possible opportunity.
> I cannot figure out if this data center panic is legit or trumped up
What recent mass-media panics have been legit?
For example, I think crypto is dumb. But I was also not dumb enough to fall for the articles like 6-8 years ago about how it was going to use more energy than half the countries in the world by 2025.
Degrowth has always been a big part of Anglophone culture. This is just the latest part of that. If you see the various parties as those downwardly mobile vs those succeeding you can automatically place people predictably into where they’ll land on various issues.
It’s all principally a tribal association. They uncritically pass around water use numbers three orders of magnitude higher than anything else and suddenly start caring about things that are a fraction of alfalfa shipments to Saudi Arabia.
> The mean air temperature on the upwind side of the facility was approximately 42.7 °C, increasing to 43.5 °C in the neighborhood near the eastern boundary of the data center campus on the downwind side. The observed ΔT of approximately 0.8 °C extended roughly 500 m downwind
Seems like we do observe heating but a much smaller average delta T than the title would hint at; I didn't see a max distance they evaluated over, seems like 500-1000m was their search window from their text. The next paragraph discusses how one of the sites they compare to is a relatively lower average temperate than the rest of Phoenix. Not really convinced by this methodology. Seems like doing something in a flat field with an array of temperature sensors would be a much more robust way to look at heating effects over some reasonable distance. Plenty of data centers fit the bill of flat and semi-isolated, curious why they chose these sites to look at
The whole data center heat/water thing is like not even in the top 100 risks/issues related to AI, but it's all people seem to care about. Why is that?
This is a pretty weird assertion to make on HN when we've had constant AI threads about everything from mass surveillance and job threats to bio-warfare agent creation. Or just simple economic debt and risk. People clearly care about all sorts of aspects. Even in mainstream media there is tons of coverage on many aspects. But if you're talking purely in terms of the law and local activism:
>Why is that?
The way we have things structured (for good and less good reasons) in the judiciary and local politics is heavily biased towards immediate or very short term, concrete, and individualized/divergent harms vs broader societal costs and risks. Water, heat, electrical access/costs, and noise all meet those standards. They can be objectively measured, impose significant immediate burdens on a small population living near the DC while little (and over time) to nothing at all elsewhere, have clearly priceable harms and/or mitigations, and are instantly graspable by everyone regardless of education/research. On a practical level, they also involve a significant degree of local power without diffusion across as many people, including traditional things like zoning. A small motivated group living there can definitely make a difference, and definitely has standing for lawsuits too.
So of course those are the first and most immediately loud focuses in that arena. The broader stuff is also under scrutiny, but at a different scale right?
Right, but even then, why would we not discuss noise pollution?
The water figure folks cite is wildly manipulated, it really confuses me why a known piece of misinformation is so routinely used. Water usage for global data center cooling is under 30B gallons of water. You only get to the 300B number by including the estimated water required for the power generation. Data centers are the only technology where we include estimated water usage for power generation under general water requirement. And even with that fudging, global water usage by data centers is a magnitude less water than just the leaky pipes in the United States.
There are much stronger points to be made without resorting to misinformation. I think points around zoning + noise pollution is something that everyone, besides our most ardent AI stan, can get behind. Every pro-ai person I know are all pretty happy to say that data center should not be built in residential zoning
There are many reasons to not like these things nearby (anywhere it affects power or water too). And on top of that they've been "somehow" getting massive incentives and tax cuts, despite bringing in very little money or employment, which costs the area even more in future years.
It's everything combined, and pushback has been somewhat successful. Success in a protest tends to help bring more success to the same protest elsewhere.
> The whole data center heat/water thing is like not even in the top 100 risks/issues related to AI, but it's all people seem to care about. Why is that?
Another commenter said the concerns about heat/water were because they were being built next to their houses, and I responded to that. I am specifically asking this person why they are touting water usage and not noise pollution as the main problem. I was not asking generically why no one talks about it
But, had I been, it is not all of these things because one of those things is abject misinformation - water usage is not a real problem and touting it as the issue hurts the protests. 30B gallons of water globally for data center cooling is 100x less than the leaky pipes in just the United States
If data centers are such a big issue that we must stop them, communicating that without lying is the protestor's responsibility. You'll never have a successful protest if it is motivated by falsifiable information and water usage is still the number one thing I see written about and spoken about my protestors. It is impossible to get on board so long as that misinformation is front and center of most arguments
It can be an issue for the community in the same watershed, as well as being affected by what they do with any waste water. There have been quite a lot of extremely-drawn-out lawsuits about them drawing more than claimed, dumping hot water into streams and damaging the ecosystem, and polluting other water sources by dumping dirty water into drinking water affecting areas.
I agree the total usage isn't particularly large, and any "they'll use up all the fresh water everywhere" FUD is flat out wrong. But it can be a large jump for an area, and they've broadly shown themselves to be incredibly amoral and willing to fight legal battles with infinite money just because they can. They are not good neighbors.
Agreed, it can. And by reserving that argument for when it is an actual problem, it strengths the overall position
> There have been quite a lot of extremely-drawn-out lawsuits about them drawing more than claimed, dumping hot water into streams and damaging the ecosystem, and polluting other water sources by dumping dirty water into drinking water affecting areas.
And I support strict EPA guidelines and heavy fees for violators! I just think that we should keep the conversation clear and not introduce facts that can be easily overturned as it moves the conversation away from impact and to the argument being made
By then it is too late, plus they never pay enough in damages. And the EPA does not have teeth currently, and damage will be done while fixing that too.
So people fight by pattern recognition: abnormally incentivized and extremely large? Kill it with fire immediately. It's mostly reasonable and more effective than anything else at the moment.
I've seen lots of talk about noise pollution with them? There's infrasound concerns and etc. No idea how real that impact is, but it's apparently measurable for a few miles away.
It mostly needs zoning as you say and maybe require each project put down a bond to supply a power plant expansion / new plant that covers their projected energy uses, imo. That sorts out a lot of issues. Maybe put some kinda tax on them and use it to pay for local infrastructure or schools for PR too
I'm not saying the noise pollution concerns aren't real, but these are often the same people who had "concerns" with 5G.
Most datacenters I've been to around the world are barely noticeable. Most of them are in warehouse districts and the AC chillers are barely noticeable over the chillers on other nearby buildings. Often, they WANT to be nondescript.
That being said, I've not been to any of these new ones which are both significantly larger and also can have onsite turbines that run all the time and not just during power outages.
> The whole data center heat/water thing is like not even in the top 100 risks/issues related to AI, but it's all people seem to care about. Why is that?
Another commenter said the concerns about heat/water were because they were being built next to their houses, and I responded to that. I am specifically asking this person why they are touting water usage and not noise pollution as the main problem. I was not asking generically why no one talks about it
"residential zoning" looks like an intentional misunderstanding of the problem. Putting one outside of "residential zoning" surely isn't a problem even if it robs an entire residential zone of power, right? Or then governments can't just re-allocate residential zones to industrial zones to get those juicy deals, right?
The way you're using language here is putting the problem so close to the edge that abuse, ignoring the quality of lives of residents, would seem just so logical.
> Or then governments can't just re-allocate residential zones to industrial zones to get those juicy deals, right?
Which is why we need laws around where data centers can and cannot be built. Something I clearly alluded to in my first comment. Do you want to make up things I never said such that you can make some snarky comment and feel good about yourself or have a real conversation? If it is the former, fuck off.
Its a proxy fight - the data centers represent AI in most peoples mind, and any animosity towards a nebulous idea of "AI" can be redirected to these concrete, physical data centers.
All that being said, if your actual goal was to hinder AI growth in some way, slowing the data center buildout in the US would probably work towards that end reasonably well.
It's possible to avoid interacting with AI in every day life but it's impossible not to be affected by second order effects. The impact from a data center boom fuelled by dirty electricity is real.
You shouldn't try to assess what people care about from what you see in the media, including social media. Focused thought and gigantic resources go into shaping the public perception of reality through media.
For example, the lower Colorado River basin is facing heavy drought, extremely low reservoirs, and federally mandated cuts in water shares. Awkward time to build water intensive infrastructure, even if the usage is relatively low.
There's a certain group of people (the kind of people who used to spend a lot of time on twitter and now spend a lot of time on bluesky) that's convinced that their environmental activism is what defeated NFT and are trying to repeat that to defeat AI because it's the same "tech bros we hate".
It's much more tangible for your average person than circular financing deals or chip supply constraints. And unfortunately social media is an amplifier for tangible misinformation.
People are terrified of AI, for a big number of both silly and serious reasons.
That fear and rage needs to be expressed somehow, ang going NIMBY on local data center construction is probably the only real way most people can do anything remotely affecting the AI boom.
Industry build-out in the USA? It's raising our energy and water costs.
Lack of industry build-out in the USA? We're losing our industrial base to the Chinese.
Social media always promotes the angry take because that's what maximizes engagement, so everything that happens (or doesn't happen) is always bad. Doesn't matter what it is. Everything is bad and we're all gonna die. Better keep scrolling to find out more!
Using energy always raises nearby temperatures since all energy becomes heat. If any facility is using 36 MW of electricity, then 36 MW of heat is going to be entering the environment somewhere. Though I think what's interesting here is that the effects of this extra heat is highly localized to urban areas.
But just trying to say this isn't data center specific. A steel mill with 36MW of electric arc furnaces is going to release the same 36MW of heat into the general environment.
Not necessarily; a building satisfied by its own rooftop solar should be roughly heat neutral since that energy was already there. The problem is with the power density these buildings imply, requiring energy movement. Everyone could run Qwen 3.8 27B at home with a solar panel and it'd make no difference. Companies could run their traditional servers on-site at their offices, etc.
Not quite. If the roof were highly reflective and it's a clear sky, it would reflect sunlight (including infrared) back into space. Of course, we're talking about such minimal levels of power here compared to, say, heating a house in the winter that it's irrelevant.
It needs around 300 W to run Qwen 3.8-27b here, which is roughly eqivalent to me as a human walking around and doing basically anything other than sitting still.
With all the data centers being built - who is using all of this compute? Am i really underestimating how demanding it is to be using ChatGPT or whatever for basic tasks?
Coding agents are increasingly popular right now, and they can use millions of reasoning tokens per hour. Basic ChatGPT queries hardly use any tokens by comparison.
Zoning is a per-city and per-county thing. Traditionally, they usually occupied some sort of commercial / light industrial zones, the same as office parks and warehouses. But hyperscale stuff often requires one-off variances / rezoning.
industry or heavy commercial, which is where some minor wrinkles have been coming up because they usually don't need special permission to build as long as the zoning fits
honestly I'm sure other industrial uses can cause temperature increases too, the nature of industrial zoning on its own tends to reduce green space which can lead to the heat island effect, so this is at least partially because hating on data centers is the issue du jour
If they're zoned like heavy industry, that seems fine for any noise/ heat etc. If they're zoned for light industry or similar to office buildings like the other comment says I feel like that could have different impacts - an office building will have some of this but probably has much less equipment and energy needs, less venting heat out maybe?
But if we're going to keep factories and chemical plants away from housing we can keep data centers over with them I think. Out of sight, out of mind, less chance of local protestors, maybe even closer to good power plants for transmission. Everyone wins?
* a bunch of places didn't have the foresight to actually put buffers between industrial and residential zoned areas (either due to space limitations and/or poor planning)... sometimes people only notice when the empty lot they lived next to for a decade actually gets something built
* currently some protests are whipping up enough fervor where location doesn't really factor, people are fighting against any data centers within city limits at all
It's the issue du jour because of foreign influence. The situation in Phoenix is due more to poor zoning since the DC is right across the street from a residential area.
If you want people to be more supportive of it, there are pretty direct ways to do that. Donate heavily to the local community to make up for impact, build new power plants for the project to pay for itself, don't put in gas powered generators, stay away from housing...
3) actual fucking nutjobs in charge of these companies who can't seem to simply not say actual nutjob shit all the fucking time
4) entry level jobs disappearing
5) the obvious fact that "I am building the most powerful technology ever" means "the already-powerful people who own these technologies and companies are working to become vastly more powerful"
And I say this as someone who uses and loves using AI heavily across both my personal and professional life. It's better to grapple directly with the fact that people have very legitimate fears and concerns around this technology and the people who own it.
Reading only the abstract: "We estimate that data centers caused average retail electricity rates to fall modestly in the United States from 2015 to 2024 using an instrumental variables approach... We caution that future supply constraints could reverse the effect."
I've only heard about datacenter protests in the past 2 years. So perhaps what the authors warned about is happening now?
Data centers directly led to my rates increasing over the last ~6 months where several came online locally without supply increasing. Natural gas prices also impacted it, but the data center buildout over the last year has not helped.
That article is two years out of date, and looking at nationwide electricity usage doesn't account for the local impacts of these data centers.
(* nb4 "it's summer": talking about price per kWhr)
> I'm sure other industrial uses can cause temperature increases too
If anything about this paper is accurate, which it clearly is not, then the >8000MW rejected to the atmosphere from the nearby Palo Verde power station should be raising the air temperature at least 250C, 1km downwind. Which it obviously does not.
Interesting, as we increase our total global power consumption, losses in the form of heat will increase proportionally.
This is more of a zoning problem
"...downwind air temperature warming as high as 2.2 °C, with average downwind air temperatures 0.7–0.9 °C warmer than corresponding upwind areas. Thermal signatures were detectable at distances up to 500 m from facility perimeters"
"The attribution of this warming to data center waste heat is supported by the consistent alignment of the temperature signal with the prevailing wind direction across multiple sites, dates, and meteorological conditions. "
Well, no? That could also mean that the heat island effect of the (statutorily mandated) employee parking lot is causal.
You can look up the NTT PH1 Datacenter from the study on Google Maps. It doesn't have much of a parking lot but due to poor industrial zoning, it's right next to a residential suburb. It's basically a failure of urban planning.
It's in the middle of a large, disturbed area with big commercial buildings, roads, and parking lots, only some of which are data centers. This type of disturbance tends to get hot in the sun. But the larger question is how they are attributing the temperature at the end of the cul de sac to the data center when the somewhat more obvious reason for a hot spot there is that nothing shades it. That is, the measured cooler streets are shaded by the houses, but the edges of the neighborhood are shaded by nothing.
Anyway the main problem with this paper is it does not have an explanation for why the response isn't linear. If areal density of heat rejection was able to raise air temperatures this amount, then nobody would be able to go anywhere near a thermal power station without being vaporized.
All I can think of when reading this was Hank Hill yelling at his neighbor: "Dale you giblet-head, we live in Texas, where it's already 110 in the summer. And if it gets one degree hotter I'm gonna kick your ass!" [1]
The referenced 169 MW location is literally across the street from a residential neighborhood, where the article indicates they measured temperature differences[0].
> This gap is significant because many facilities are sited adjacent to residential neighborhoods (Fig. 1), and their air-cooled condenser arrays discharge air at temperatures 8–14 °C above ambient—often exceeding 50 °C during Phoenix summers—with air velocities of 2–4 m/s [15,16], creating thermal plumes that are advected downwind over inhabited areas. For example, the Iron Mountain Data Center (Fig. 1(c)) has cooling equipment located less than 50 m from the nearest three-story apartment building.
That doesn't make the slightest difference for two reasons:
1. Natural gas still produces CO2, which is a climate change driver. (And, since burning more of it will only increase global temperatures, the data center will have more trouble losing its excess heat.
2. Even if a data centre uses a totally clean energy source, it's still excess usage. That clean energy could have been used elsewhere. Until all energy is clean, usage by unnecessary sources is a net negative.
that's already raising the local area temps, why not put that wasted fuel to work instead?
build the datacenters literally next to the fracking fields, absolutely no-one there to complain, they all moved away already from the toxic air/water and there's already unlimited fuel
> Crusoe has built their business around turning flaring from an operational and regulatory risk for oil and gas companies, as well as an existential risk for our planet, into a multi-solve solution. Crusoe doesn’t just mitigate flaring emissions. They capture the gas, convert it into energy, and use it to power computing infrastructure.
wow Crusoe deserves more press if everything they are saying checks out
flaring has upset me for a decade, even if it was just used for bitcoin mining it would have at least done something vaguely beneficial compared to just raising earth temperature
but if it can help solve the "AI" energy demand boom that would be huge
because it's not like this country is ever going to stop fracking at $80-$100 a barrel now
Methane has a stronger greenhouse effect than CO2 (until it decays into CO2) so it's probably better to burn it and do no work than just vent it, and better still to leave it in the ground.
But ... but ... but ... ai companies have safety teams that ... well ... make sure some kind of singularity mumbo jumbo threat wont happen. The thing they dont care about are ... well ... actual people and actual environment.
I cannot figure out if this data center panic is legit or trumped up. There are legitimate nation-state and political entities that would love to create pressure for degrowth in this area. And, pollution and natural resources are a concern, yes, fine, sure. NIMBY happens, I get it ...
*BUT*, we've been building huge data centers for decades and they are minuscule compared to the size of the earth. My skeptic meter maxed out when i saw a Tesla with a "End Data Centers" bumper sticker on it. This is just the latest popular panic, isn't it?
We’re building data centers in the U.S. at an unprecedented and increasing rate [1]. Is it so hard to imagine that something which was fine at one scale may begin to cause issues at some massively higher scale?
If I drink one glass of wine per day, it’s probably fine. If I drink three bottles per day, I’m gonna start having problems.
[1] https://ourworldindata.org/grapher/monthly-spending-data-cen...
Yeah, but when we account for that scale, and look at resource usage, datacenters aren't topping the charts, or even really getting in the top 5, for anything.
You have people who eat meat at every meal losing their mind over water usage and people who set their AC to 68 panicking over energy usage.
Eating meat fills my belly. Setting my AC to 68 keeps me cool. A data center costs my community money, damages the environment directly and indirectly, and further enriches a few already rich people.
Can you spot the difference?
AC at 68F is obnoxious and wasteful. At least data centres enable me to communicate with distant colleagues, watch enriching movies, and use very strong AI tools to accelerate my research.
People say this and then use AI for literally everything they do.
Valid point, but this looks like (roughly) 1 billion to 5 billion in a period of 10 years.
5x in a decade. The way it’s talked about you would think it’s orders of magnitude.
Your point would stand better without the insincere analogy.
Insincere analogy. F-ing lol. You AI Slopologists are all the same. Can’t see the forest from the trees. Or was that insincere as well? :eye roll:
Complete tangent, but my impression is that one glass of wine per day is "fine", but you're still statistically better off in a non-trivial way by going down to zero. Not counting the benefits of course.... :)
To continue to take us off topic, there is actually pretty limited evidence that this is true.
It is true that wine doesn't do anyone any good (unless you count it making people more social or something). But the effects of one glass a night is indistinguishable from zero. There are definitely statistically significant morbidity effects at 3.
I've been shocked reading public health proclamations that "anything above zero" is bad, and then reading the actual underlying paper they refer to, which clearly states exactly the opposite finding. I think it is an extension of the public policy belief that the public cannot be trusted with moderation, and so therefore must be communicated to in absolutes, and nobody will read the underlying paper anyway.
Some of it's a panic, but I'm leaning towards many of the concerns being true or at least partially true due to the sheer scale of the construction. The figures I've been seeing are something like half of all current private construction projects are data center related these days and it's the biggest single sector of construction in terms of money spent.
It's also a lot easier to protest data centers then AI companies and they generally provide minimal benefits to the local communities (a brief job boom for construction but minimal reoccurring benefits afterwards) and quite a few negatives like power consumption and noise. People can't stop their jobs from replacing them with an LLM, but they can at least stop the local data center from being built and raising the electricity bill by $70 a month[0]. This batch of data center construction is essentially yet another physical representation of the (often negative) changes the general populace faces with AI.
[0] $70/month is how much my old power company said they'd have to raise prices if several proposed data centers were built. I was already paying $150-200 a month to power a 1 bed apartment so you can bet your ass I filed a complaint with the county about it.
> they generally provide minimal benefits to the local communities
Data centers pay Loudoun County, Virginia $26 in taxes for every $1 in services they consume — nearly $1.3 billion, 45% of the county's budget — which is why Loudoun homeowners pay the lowest property tax rate in Northern Virginia.
https://www.loudoun.gov/faq.aspx?TID=241
What I don't understand is why communities want to block the data centers.
Just make them pay!
Your electricity price shouldn't go up because of them. So just make a rate card for data centers that ensures the burden doesn't fall on residents. That's what Oregon started doing, and residents' electricity prices actually started falling, because the fixed costs of energy production were being spread across more kilowatt-hours and the data centers weren't getting a special deal.
I've also seen counties offer tax incentives. Why? Just have them pay property tax. Then you get nicer schools, more police, better roads.
Instead we have these campaigns to stop them, instead of making them a local cash cow. Loudon County gets half of all its revenue from data center property tax. Its residents were already rich, and now their local taxes are artificially low. Who wouldn't want that?
They do pay, that's why these towns are doing all these shady things to minimize public input.
The problem is that the towns generally want to spend the money on new stuff rather than give the money back in tax cuts.
> What I don't understand is why communities want to block the data centers. Just make them pay!
What kind of question is that? Communities that don't want them?
You think money can fix and buy everything?
Besides, they pay already.
Wait… large customers move in and your rate goes up?
Forget about it. These AI companies could really help themselves by subsidizing local rates. Imagine if a large AI datacenter moved in AND your rate dropped by $70!
Everyone would be singing their praises.
Oregon just passed a law saying data centers will pay more for electricity than normal people, and the math says that everyone's rates will be going down.
So, again, I think it depends on whether you live in a red or blue state.
https://www.opb.org/article/2026/06/04/data-centers-utility-...
>Wait… large customers move in and your rate goes up?
demand goes up + supply remains steady = price goes up.
Also infrastructure costs might be billed differently between industrial vs retail.
(And the marginal cost on infrastructure isn't necessarily paid be the people increasing the demand)
Well they say
*As high as 2.2 °C, with average downwind air temperatures 0.7–0.9 °C warmer than corresponding upwind areas. Thermal signatures were detectable at distances up to 500 m"
"As high as", combined with "up to" show two upper limits, but you can be sure if they were simultaneous they would have said so, especially when considering their prhasing of downwind being warmer comes after mentioning the upper limit. It really seems like you are supposed to overlook that it's an upper limit and subconsciously add another 0.7-0.9 on top.
What they are saying is you can feel the warm breeze coming off the hot thing if you stand near enough.
A comparison to other industrial sites would put it in perspective. As would a distinction between localised warming which is what is happening here compared to global warming where small temperature differences at a global scale are far more significant. The publicity about global warming risk has put a lot of effort attuning people to the significance of those small temperature shifts, to co-opt that education to make people fear a situation where it doesn't apply is undermining their work. That puts us at more risk than any one data centre. People can observe a local increase of 2 degrees in relative comfort and think that it won't be catastrophic at a global scLe.
Seems to me that the problem isn't that the data centres are being built. It's that the people building them aren't paying for their full cost. Despite the organisations building them being some of the richest in the world that can absolutely afford to do so.
(externalities in terms of environmental impact and resource usage (water, electricity, etc) usage driving up costs for everyone else).
I'm a hobbyist IT enthusiast, not a professional computer guy, but from my point of view there's a big difference between a bunch of servers hosting something people want, like Netflix, and a warehouse full of GPUs running around the clock to make Lobster Jesus or whatever.
It's also a proxy for people's feelings on AI--essentially the only way people have to fight back against the new technology that makes eternal garbage and is supposedly coming for their jobs is to push for banning "data centers."
Totally agree that it's just a proxy for people's feelings on AI, but my Lobster Jesus is much more important to me than a Netflix show! All of the datacenter usage (for inference) is just serving something people want. Time has already shown that the datacenter for training is also just creating something people want.
Lobster Jesus FTW.
How is there a big difference between those two things? People want Netflix, people also want the output of AI whether that is lobster Jesus or something else. This seems more like substituting what people want, which clearly includes AI, for what some people think others should want. I can guarantee you there are people who don’t want Netflix and think it’s a waste of resources that encourage people to binge TV, become a couch potato, and end up being drags on the healthcare system. Should their preferences override those who do want it?
Democracy, even at its most ideal, is often about overriding preferences. Netflix helps to float an entire industry of creators and a tangible pop culture. It's not entirely clear (or softened, maybe not clear to communities) what AI build out floats currently except AI companies and the tangible negatives of generative AI.
I think that potentially the difference is that, while people do in some sense want AI, they don't want it in the same way that they want Netflix. I'd compare it to Amazon: it perfectly possible for people to like the ability to like 2 day shipping while also liking that the site is flooded in cheap crap.
So maybe it is that people like AI in certain domains but HATE the way that it is being used overall?
> How is there a big difference between those two things?
Netflix just serves existing data so their biggest need is data storage ala CDN for closer retrieval to the user. Very low requirements. Creating sweet baby lobster jesus requires incredible amounts of computer/power/cooling. The requirements are totally different. Yes, both ultimately deliver pixels to consumer's eyeballs, but how they get there are so different that it's laughable that people cannot see the difference.
There's also different building pressures. Before, companies would want to build / power / maintain the smallest amount of data center hardware they could while still serving their needs. Overprovisioning just meant worse opex number.
Now, because supply is so constrained relative to demand, there's no limit. More is always better, which leads to a crazy amount of construction (as its negative effects).
> there's a big difference between a bunch of servers hosting something people want, like Netflix, and a warehouse full of GPUs running around the clock to make Lobster Jesus or whatever.
ChatGPT has been the most downloaded app every month for a year and a half.
It's gotta be something. The actual apps I use to do work on my phone either were included in the OS or downloaded long ago.
The conflating of a data center and an AI/GPU farm is disingenuous at best, and the AI people know what they are doing. None of the "traditional" data centers are being powered on site with gas turbine generators. New GPU farms are also being designed with much larger footprints with much more needs from the public utilities.
AI inference and training GPUs require way, way more power than anything manufactured before. Like, between 2020 and 2026 we went from ~400W for an A100 to ~1400W for an NVIDIA B300. And you cannot make that happen without changing the datacenter itself. More noise, way more complicated cooling setup, new power sources, etc. Not all datacenters are created equal.
The financing for datacenters is also very shady and its own can of worms
> The financing for datacenters is also very shady and its own can of worms
Could you expand on that?
I don't necessarily know what the parent was referring to, but I can see two this:
Megacorps often get significant tax breaks to build AI in a community.
Megacorps advertise that datacenters will create jobs but don't mention that most of those jobs will disappear once construction is finished.
I don't think it's necessarily shady, but let's say that we're assisting at bets and capex spending we've never seen in history.
Companies like Meta, Alphabet, etc that were drowning in cash flow and buying back their stock are now back at issuing shares, creating a huge number of SPVs, issuing debt, etc.
There's a gigantic spending spirale with no signs of ending. Nvidia is financing companies to buy its own hardware.
Mind you, this is nothing new in some businesses. Airbus and Boeing have been lending money to companies to buy their own airplanes forever.
But its just all too much, with very little money coming. To understand how massive is the spending: even if every adult on the planet spent 250$ on AI per year this would still not cover anywhere near what's being invested.
Just because we've been doing something forever does not mean that a recent resurgence in policy interest and a re-evaulation of the social worth of these projects is a waste of time.
Part of the problem is AI datacenters make bad neighbors. Because they are in such a rush to become operational they cut corners.
Instead of AC they use swamp coolers draining aquifers and thus wells of the neighbors.
Instead of using new power plants (AKA using renewables as most new plants do, and all old datacenters did). They either hook up to the grid and jack up everyones electric prices or use generators because they can't be bothered to wait for a grid connection.
All of which is to say while on the grand scale they may be fine. Locally they can be unpleasant to have in your back yard.
If the LLM datacenter builders and operators weren't cutting corners and rushing and doing everything they can to be good neighbors. There would be much less push back.
I say this as someone with 2 huge pre-AI datacenters in my town that have been fine.
That's just a non sequitur.
Compare the scale of datacenter construction in the past decades vs the pace post 2022.
Also the general computing datacenters of the past are vastly different in societal utility vs the GPU only AI datacenters, which do just one thing
It's just the outrage story of the month. There will be some new moral panic in September and we'll all move on. The attention span for topics like this on social media is quite short.
The risk is that reacting to the panic, states lock in laws with destructive moratoriums they politically can't back out of later.
> There are legitimate nation-state and political entities that would love to create pressure for degrowth in this area.
This is the trumped up rhetoric you speak of.
Small scale natural gas turbine plants are bad for local human air due to NOx emissions.
100% true yet even when you solve this (for example with clean hydro in the northwest), you make zero progress with the anti-datacenter people.
Right, as with almost everything with people, it's an emotional reaction (fear) with a thin layer of rationalization on top.
America is huge. Why do they need to be constructed next to people's homes?
That's where the resources are. You wouldn't expect some uber wealthy billion dollar company to have to provide power/water needs way out in the middle of nowhere would you?
Is there a better alternative for on-demand, on-location power generation?
One strange form that the post-truth era takes is deciding that stances you don't believe in aren't even real. Saves you the trouble of having to engage with them meaningfully.
[delayed]
These days there always seems to be panic regardless. I find it better to look at the merits of individual claims rather than try to group everything together into a binary answer like a possible conspiracy on one thing means nothing else needs to be considered.
Zoning, grid strain (->pricing), and chip strain (->pricing) seem to be pretty legit most of the time I look into them. Noise seems to have a lot of "they opened a 5G tower in the town and now I get nightmares" type panic but also some legitimate problems which, ultimately, relate to zoning. I try to reserve NIMBY for "things I want in someone's back yard, just not mine". The reality is data centers don't fit that because there is no reason they need to be in anyone's back yard. They are perfectly happy to be built in the middle of other industry/commercial blocks or even in the middle of nowhere.
I think if the grid had been better prepared for massive demand growth with cleaner sources than before and things came in slow enough that zoning didn't feel rushed people wouldn't really make as much of a stink around it. Unfortunately, the US spent many years with flat grid demand and companies want to cram these DCs in as fast as possible because their demand was for last year already.
For some large factor, it's also a reason to really hate on AI. Which there may well be valid reasons for, I just also wonder how many of these people actually care about the environment and how many are just saying it because it's another con to list.
My great-granddad was unconcerned about the deforestation of the Amazon during the 1970s until the little forest on the way out of his small town was cut down for a new subdivision.
One HUGE issue is that datacenters don't have an afterlife in the communities where they're located.
They add to the Rust Belt once their lifetimes end, contrast this with old brick warehouses which were eventually repurposed as homes and offices.
If they were built with a bit of thought around what happens once they depreciate completely and they're abandoned/sold for residual value they wouldn't be as despised.
Current construction completely ignores the nearby communities present and future development.
If the ai bubble bursts, these buildings will be able to be repurposed unlike rust belt factories. With all of that cooling, you could turn them into a weird form of multi-family housing. Or use them for indoor skate pars, kart racing tracks, whatever you can imagine. I don't think rust belt is an appropriate comparison. Also, these places don't hire that many people like a factory did. If the factory shut down, entire towns could collapse with it. If a data center shut down, maybe 20 people would notice????
> Current construction completely ignores the nearby communities present and future development.
The issue is an incentive one, for the companies involved in the construction.
Right now, they are getting paid so much to build it fast. Time is of the greatest importance. Nothing else is mattering to anyone involved except how fast datacenters are being built. At best, this is an after-thought.
As such, I believe that we have to change the incentives to create more sustainability but it requires time, discussion. Perhaps laws can help in making them more sustainable but currently, the sad reality is that for all the parties actually involved, I don't think that many truly care about the after-math or repurcussions of it as much as its the only thing that they want right now
AI datacenters are a completely different beast than the "normal" cloud ones.
They use order of magnitudes more resources.
It’s not so binary
Basically, anything people decide not to like is targeted for its climate impact
Some of that targeting is valid and points out optimizations that can be done
They want you to ignore their air conditioning and fridge’s greater relative impact, because the thing they dont like has “no use” in comparison to the thing they do like. They also believe that any power use is easily transportable to the use they do like, which is usually false.
But there are configurations of data centers with less environmental impact, which should become ubiquitous
The outrage is real to the individual, but manufactured. Media (e.g. 404media, Ed Zitron) promotes anything that engages for clicks, views, and subscriptions. Other parties do it to hinder competition, weaken the US, or distract from much worse polluters.
It kind of feels to me like a big element of this is that the social contract in America has broken down. Why should an individual accept any sacrifice whatsoever if all of the profit is going to go to the billionaires who own it? If they do get built and someone suffers an externality, why would they believe that society will do anything to make things right?
Basically, I think that the the capitalist class are finding out that things like social safety nets, socialized healthcare, income equality, etc. aren't just important for humanitarian reasons; they are also prerequisites for people to trust society enough to be willing to coordinate around major projects.
> Why should an individual accept any sacrifice whatsoever if all of the profit is going to go to the billionaires who own it?
At some point, their 401Ks are going to depend on it.
I think that's another example of the social contract breaking down. Lots of people either cannot afford to invest or no longer believe that investing will allow them to afford things like a house or retirement that were traditionally part of being middle class.
I imagine that things would be dramatically different if most people believed that the success of data centers would directly lead to their economic success.
I've done a bit of research into it, and I think it's a mix.
The energy thing is exaggerated. Want to compensate for a week of heavy AI coding usage? If you drive to work, telecommute for one day. Skip one long-distance DoorDash order. Cars use tons of energy, a lot more per-user than the fraction of data center capacity you're using with your AI stuff. A burrito on DoorDash delivered using an average MPG gas car uses something like a week's worth of heavy AI usage in energy.
The water thing is region-specific and where is a problem it's due to incompetent or corrupt goveranance. Bottom line is that if a data center is using too much water, it means they're not paying enough for that water. Make them pay more for their water and they'll magically discover closed cycle cooling.
The noise thing is in some cases real, and needs to be fixed with local regulation or a class action suit by neighbors to force them to deal with the noise problem.
I am a little puzzled though by where these things are being built. Phoenix? Other than cheap solar power I can't tell why you'd want a data center there given the energy needed for cooling.
Why are we not putting them up in the Arctic circle? Build run-of-the-river hydro in northern Canada and put them way up there. Free cooling, especially in winter. AI is not a very latency-sensitive load so an extra 50ms doesn't mean much.
I suspect state incentives are a reason. The giant Amazon data center proposal in Texas with the giant natural gas power plant sounds a lot like "taxpayer funded indirect fossil fuel industry subsidy." Offer some kind of big state incentive to get a huge natural gas customer with long-term locked-in demand.
> The energy thing is exaggerated.
https://www.wpr.org/news/wisconsin-energy-demand-to-rise-dri...
> Wisconsin energy demand to rise 40 percent by 2032, driven by data centers, draft report finds
> Draft report: Over 72% of peak electricity load is 'associated with three active data center developments'
https://www.wpr.org/news/we-energies-add-3gw-electric-grid-d...
https://www.cleanwisconsin.org/wisconsin-ai-data-center-powe...
> Two AI Data Centers Require More Energy Than All Wisconsin Homes Combined: Just two AI data center projects — the Microsoft data center in Mt. Pleasant and Vantage data center in Port Washington — will require a combined 3.9 gigawatts (GW) of electric power, which is enough energy to power 4.3 million Wisconsin homes. There are only 2.8 million housing units in the entire state.
https://www.ladwp.com/who-we-are/power-system
> In fiscal year 2023-24, we supplied more than 20,749 gigawatt-hours (GWh) of power for more than 1.6 million electric service customers, including nearly 6,000 in the Owens Valley.
That's an average of 2.3 GW over the year.
-----
> The water thing is region-specific and where is a problem it's due to incompetent or corrupt goveranance. Bottom line is that if a data center is using too much water, it means they're not paying enough for that water.
https://www.opb.org/article/2026/01/15/as-googles-water-dema...
> When Google purchased a defunct aluminum smelter in The Dalles, it also acquired the right to pump massive amounts of water from underground — up to 3.88 million gallons a day. That’s almost as much water as the entire city averaged per day in 2024.
> The corporation transferred those water rights to the city in 2021, Bentz noted, allowing The Dalles to pump more water from underground. The city only obtained part, not all, of the former smelter’s water rights, Mays said after OPB’s story published.
It's the rights to pump the water. Google can pump up to 3.88 million gallons per day and the city can't bill them a penny for that. Also, because it's an old claim, it's likely more senior to other water rights.
https://www.oregon.gov/owrd/programs/waterrights/pages/defau...
> Make them pay more for their water and they'll magically discover closed cycle cooling.
A closed cycle cooling system doesn't solve the overall problem of water usage. Yes you have a cycle within the datacenter, but you need to then move that energy somewhere else, out of the system. Eventually that energy needs to find its way in the environment. A close loop system for a datacenter reduces the water use on site, but actually increase electro use. If that electricity comes from a power plant that is cooled by water you just moved the problem upstream
That's true. Making them pay enough for water might mean the data center doesn't even get built there. It gets built somewhere more logical with abundant cheap water.
It's gotta be costs, right? Everyone wants this done fast and at large scale, so they're going for any sites with existing large electrical connections I assume.
It is 100% deranged nonsense. These people have absolutely lost their minds.
TikTok and social media
Of course, that's why the US wanted to buy it.
It's more like, not having Trump, and controlling the complete manufacturing pipeline from beginning to end
It is real but it's about the unaccountability, dishonesty, and misanthropy of corporate overlords, rather than about the externalities of data centers per se. See https://news.ycombinator.com/item?id=49180901
It was already a huge problem that powerful people in this world work to enrich themselves with no regard to regular people's wellbeing. The rage about data centers is because it suddenly started happening in everyone's backyard, making the abstract problem into a physically palpable one.
Imo their anger is completely justified. This only gets any better when there is an actual imperical shift towards behaving prosocially and away from mindless profiteering, and people have every right to want that from people in power. I fully expect the path we're on to end in some kind of revolution if nothing changes and there seems to be no sign of anything changing.
In particular, shame on anyone who uses arguments of the form: "it's these company's jobs to maximize profit, therefore they should not be beholden to behaving prosocially; the abstract force of economic growth justifies all iniquities because it will miraculously benefit everyone eventually". This argument is a lie on every count: it's only their job because we've chosen for it to be their job; the force of economic growth famously hurts people all the time; there is no evidence that it will fix everything versus just accelerate burning everything down (quite literally in this case); and finally, people's happiness is proportional more to their ability to trust those in power to care about them and their communities than to their material wealth per se, so trust is the thing they want maximized, not profit--and they are presently starved of it, because the people with power behave like lying manipulative scumbags at every possible opportunity.
> I cannot figure out if this data center panic is legit or trumped up
What recent mass-media panics have been legit?
For example, I think crypto is dumb. But I was also not dumb enough to fall for the articles like 6-8 years ago about how it was going to use more energy than half the countries in the world by 2025.
Degrowth has always been a big part of Anglophone culture. This is just the latest part of that. If you see the various parties as those downwardly mobile vs those succeeding you can automatically place people predictably into where they’ll land on various issues.
It’s all principally a tribal association. They uncritically pass around water use numbers three orders of magnitude higher than anything else and suddenly start caring about things that are a fraction of alfalfa shipments to Saudi Arabia.
> The mean air temperature on the upwind side of the facility was approximately 42.7 °C, increasing to 43.5 °C in the neighborhood near the eastern boundary of the data center campus on the downwind side. The observed ΔT of approximately 0.8 °C extended roughly 500 m downwind
Seems like we do observe heating but a much smaller average delta T than the title would hint at; I didn't see a max distance they evaluated over, seems like 500-1000m was their search window from their text. The next paragraph discusses how one of the sites they compare to is a relatively lower average temperate than the rest of Phoenix. Not really convinced by this methodology. Seems like doing something in a flat field with an array of temperature sensors would be a much more robust way to look at heating effects over some reasonable distance. Plenty of data centers fit the bill of flat and semi-isolated, curious why they chose these sites to look at
The title is correct. You are looking at average temperature rise. The article states by up to which is the maximum temperature rise.
If you catch me on fire my average temperature rise doesn't go up very much, but I have the barest of hints that it might have an impact.
Never said anything about correctness.
If only we had the same amount of concern about oil refineries and gas stations.
https://environmentalintegrity.org/news/13-oil-refineries-in...
https://www.theenergymix.com/gas-stations-leave-behind-toxic...
Fossils use, production and trade needs to be rapidly ramped down to mitigate the climate catastrophe and resulting megadeaths on all fronts.
Won't someone think of the shareholders of the oil conglomerates?
Fuck that.
Let 'em burn.
The whole data center heat/water thing is like not even in the top 100 risks/issues related to AI, but it's all people seem to care about. Why is that?
>but it's all people seem to care about.
This is a pretty weird assertion to make on HN when we've had constant AI threads about everything from mass surveillance and job threats to bio-warfare agent creation. Or just simple economic debt and risk. People clearly care about all sorts of aspects. Even in mainstream media there is tons of coverage on many aspects. But if you're talking purely in terms of the law and local activism:
>Why is that?
The way we have things structured (for good and less good reasons) in the judiciary and local politics is heavily biased towards immediate or very short term, concrete, and individualized/divergent harms vs broader societal costs and risks. Water, heat, electrical access/costs, and noise all meet those standards. They can be objectively measured, impose significant immediate burdens on a small population living near the DC while little (and over time) to nothing at all elsewhere, have clearly priceable harms and/or mitigations, and are instantly graspable by everyone regardless of education/research. On a practical level, they also involve a significant degree of local power without diffusion across as many people, including traditional things like zoning. A small motivated group living there can definitely make a difference, and definitely has standing for lawsuits too.
So of course those are the first and most immediately loud focuses in that arena. The broader stuff is also under scrutiny, but at a different scale right?
Because these new data centers were built near people's homes.
Right, but even then, why would we not discuss noise pollution?
The water figure folks cite is wildly manipulated, it really confuses me why a known piece of misinformation is so routinely used. Water usage for global data center cooling is under 30B gallons of water. You only get to the 300B number by including the estimated water required for the power generation. Data centers are the only technology where we include estimated water usage for power generation under general water requirement. And even with that fudging, global water usage by data centers is a magnitude less water than just the leaky pipes in the United States.
There are much stronger points to be made without resorting to misinformation. I think points around zoning + noise pollution is something that everyone, besides our most ardent AI stan, can get behind. Every pro-ai person I know are all pretty happy to say that data center should not be built in residential zoning
Noise pollution is definitely being discussed: https://www.usnews.com/news/national-news/articles/2026-04-2...
There are many reasons to not like these things nearby (anywhere it affects power or water too). And on top of that they've been "somehow" getting massive incentives and tax cuts, despite bringing in very little money or employment, which costs the area even more in future years.
It's everything combined, and pushback has been somewhat successful. Success in a protest tends to help bring more success to the same protest elsewhere.
> The whole data center heat/water thing is like not even in the top 100 risks/issues related to AI, but it's all people seem to care about. Why is that?
Another commenter said the concerns about heat/water were because they were being built next to their houses, and I responded to that. I am specifically asking this person why they are touting water usage and not noise pollution as the main problem. I was not asking generically why no one talks about it
But, had I been, it is not all of these things because one of those things is abject misinformation - water usage is not a real problem and touting it as the issue hurts the protests. 30B gallons of water globally for data center cooling is 100x less than the leaky pipes in just the United States
If data centers are such a big issue that we must stop them, communicating that without lying is the protestor's responsibility. You'll never have a successful protest if it is motivated by falsifiable information and water usage is still the number one thing I see written about and spoken about my protestors. It is impossible to get on board so long as that misinformation is front and center of most arguments
It can be an issue for the community in the same watershed, as well as being affected by what they do with any waste water. There have been quite a lot of extremely-drawn-out lawsuits about them drawing more than claimed, dumping hot water into streams and damaging the ecosystem, and polluting other water sources by dumping dirty water into drinking water affecting areas.
I agree the total usage isn't particularly large, and any "they'll use up all the fresh water everywhere" FUD is flat out wrong. But it can be a large jump for an area, and they've broadly shown themselves to be incredibly amoral and willing to fight legal battles with infinite money just because they can. They are not good neighbors.
Agreed, it can. And by reserving that argument for when it is an actual problem, it strengths the overall position
> There have been quite a lot of extremely-drawn-out lawsuits about them drawing more than claimed, dumping hot water into streams and damaging the ecosystem, and polluting other water sources by dumping dirty water into drinking water affecting areas.
And I support strict EPA guidelines and heavy fees for violators! I just think that we should keep the conversation clear and not introduce facts that can be easily overturned as it moves the conversation away from impact and to the argument being made
By then it is too late, plus they never pay enough in damages. And the EPA does not have teeth currently, and damage will be done while fixing that too.
So people fight by pattern recognition: abnormally incentivized and extremely large? Kill it with fire immediately. It's mostly reasonable and more effective than anything else at the moment.
I've seen lots of talk about noise pollution with them? There's infrasound concerns and etc. No idea how real that impact is, but it's apparently measurable for a few miles away.
It mostly needs zoning as you say and maybe require each project put down a bond to supply a power plant expansion / new plant that covers their projected energy uses, imo. That sorts out a lot of issues. Maybe put some kinda tax on them and use it to pay for local infrastructure or schools for PR too
I'm not saying the noise pollution concerns aren't real, but these are often the same people who had "concerns" with 5G.
Most datacenters I've been to around the world are barely noticeable. Most of them are in warehouse districts and the AC chillers are barely noticeable over the chillers on other nearby buildings. Often, they WANT to be nondescript.
That being said, I've not been to any of these new ones which are both significantly larger and also can have onsite turbines that run all the time and not just during power outages.
> The whole data center heat/water thing is like not even in the top 100 risks/issues related to AI, but it's all people seem to care about. Why is that?
Another commenter said the concerns about heat/water were because they were being built next to their houses, and I responded to that. I am specifically asking this person why they are touting water usage and not noise pollution as the main problem. I was not asking generically why no one talks about it
"residential zoning" looks like an intentional misunderstanding of the problem. Putting one outside of "residential zoning" surely isn't a problem even if it robs an entire residential zone of power, right? Or then governments can't just re-allocate residential zones to industrial zones to get those juicy deals, right?
The way you're using language here is putting the problem so close to the edge that abuse, ignoring the quality of lives of residents, would seem just so logical.
> Or then governments can't just re-allocate residential zones to industrial zones to get those juicy deals, right?
Which is why we need laws around where data centers can and cannot be built. Something I clearly alluded to in my first comment. Do you want to make up things I never said such that you can make some snarky comment and feel good about yourself or have a real conversation? If it is the former, fuck off.
Its a proxy fight - the data centers represent AI in most peoples mind, and any animosity towards a nebulous idea of "AI" can be redirected to these concrete, physical data centers.
All that being said, if your actual goal was to hinder AI growth in some way, slowing the data center buildout in the US would probably work towards that end reasonably well.
It's possible to avoid interacting with AI in every day life but it's impossible not to be affected by second order effects. The impact from a data center boom fuelled by dirty electricity is real.
> it's all people seem to care about.
You shouldn't try to assess what people care about from what you see in the media, including social media. Focused thought and gigantic resources go into shaping the public perception of reality through media.
The impact is highly geography dependent.
For example, the lower Colorado River basin is facing heavy drought, extremely low reservoirs, and federally mandated cuts in water shares. Awkward time to build water intensive infrastructure, even if the usage is relatively low.
Pretty sure everyone cares about their power bills and water bills going up. The people living next to them care about this other stuff.
There's a certain group of people (the kind of people who used to spend a lot of time on twitter and now spend a lot of time on bluesky) that's convinced that their environmental activism is what defeated NFT and are trying to repeat that to defeat AI because it's the same "tech bros we hate".
It's much more tangible for your average person than circular financing deals or chip supply constraints. And unfortunately social media is an amplifier for tangible misinformation.
People are terrified of AI, for a big number of both silly and serious reasons.
That fear and rage needs to be expressed somehow, ang going NIMBY on local data center construction is probably the only real way most people can do anything remotely affecting the AI boom.
Industry build-out in the USA? It's raising our energy and water costs.
Lack of industry build-out in the USA? We're losing our industrial base to the Chinese.
Social media always promotes the angry take because that's what maximizes engagement, so everything that happens (or doesn't happen) is always bad. Doesn't matter what it is. Everything is bad and we're all gonna die. Better keep scrolling to find out more!
Because they are hitting people's water and electricity bills (and or ability to get either at all) RIGHT NOW
Yea decreasing them
Using energy always raises nearby temperatures since all energy becomes heat. If any facility is using 36 MW of electricity, then 36 MW of heat is going to be entering the environment somewhere. Though I think what's interesting here is that the effects of this extra heat is highly localized to urban areas.
But just trying to say this isn't data center specific. A steel mill with 36MW of electric arc furnaces is going to release the same 36MW of heat into the general environment.
Not necessarily; a building satisfied by its own rooftop solar should be roughly heat neutral since that energy was already there. The problem is with the power density these buildings imply, requiring energy movement. Everyone could run Qwen 3.8 27B at home with a solar panel and it'd make no difference. Companies could run their traditional servers on-site at their offices, etc.
Not quite. If the roof were highly reflective and it's a clear sky, it would reflect sunlight (including infrared) back into space. Of course, we're talking about such minimal levels of power here compared to, say, heating a house in the winter that it's irrelevant.
It needs around 300 W to run Qwen 3.8-27b here, which is roughly eqivalent to me as a human walking around and doing basically anything other than sitting still.
“Everyone in McKinney is dead”: https://www.youtube.com/watch?v=EzA8DcAiijc
My first real job was in McKinney, sad to hear they are all dead.
With all the data centers being built - who is using all of this compute? Am i really underestimating how demanding it is to be using ChatGPT or whatever for basic tasks?
The benefit to effort ratio seems fishy to me.
Coding agents are increasingly popular right now, and they can use millions of reasoning tokens per hour. Basic ChatGPT queries hardly use any tokens by comparison.
4 degrees Fahrenheit:
> "... air temperature warming as high as 2.2 °C ..."
A modern data center (1GW) is about the same as 140k additional cars in traffic in terms of direct heating.
Seems more likely that the heating is climate change driven primarily. But, 140k cars 24/7 is a lot of cars, so???
Out of curiosity, how are large data centers zoned?
Zoning is a per-city and per-county thing. Traditionally, they usually occupied some sort of commercial / light industrial zones, the same as office parks and warehouses. But hyperscale stuff often requires one-off variances / rezoning.
industry or heavy commercial, which is where some minor wrinkles have been coming up because they usually don't need special permission to build as long as the zoning fits
honestly I'm sure other industrial uses can cause temperature increases too, the nature of industrial zoning on its own tends to reduce green space which can lead to the heat island effect, so this is at least partially because hating on data centers is the issue du jour
If they're zoned like heavy industry, that seems fine for any noise/ heat etc. If they're zoned for light industry or similar to office buildings like the other comment says I feel like that could have different impacts - an office building will have some of this but probably has much less equipment and energy needs, less venting heat out maybe?
But if we're going to keep factories and chemical plants away from housing we can keep data centers over with them I think. Out of sight, out of mind, less chance of local protestors, maybe even closer to good power plants for transmission. Everyone wins?
a couple things to note that complicate it:
* a bunch of places didn't have the foresight to actually put buffers between industrial and residential zoned areas (either due to space limitations and/or poor planning)... sometimes people only notice when the empty lot they lived next to for a decade actually gets something built
* currently some protests are whipping up enough fervor where location doesn't really factor, people are fighting against any data centers within city limits at all
It's the issue du jour because of foreign influence. The situation in Phoenix is due more to poor zoning since the DC is right across the street from a residential area.
https://www.nytimes.com/2026/07/09/business/china-russia-ai-...
If you want people to be more supportive of it, there are pretty direct ways to do that. Donate heavily to the local community to make up for impact, build new power plants for the project to pay for itself, don't put in gas powered generators, stay away from housing...
Or because people see:
1) their electricity bills rising
2) their land prices (read: rent) rising
3) actual fucking nutjobs in charge of these companies who can't seem to simply not say actual nutjob shit all the fucking time
4) entry level jobs disappearing
5) the obvious fact that "I am building the most powerful technology ever" means "the already-powerful people who own these technologies and companies are working to become vastly more powerful"
And I say this as someone who uses and loves using AI heavily across both my personal and professional life. It's better to grapple directly with the fact that people have very legitimate fears and concerns around this technology and the people who own it.
The Chinese and Russians didn't double my electricity rates this year.
Neither did the data centers. Datacenters use less than 5% of US electricity.
https://arxiv.org/abs/2606.19777
Reading only the abstract: "We estimate that data centers caused average retail electricity rates to fall modestly in the United States from 2015 to 2024 using an instrumental variables approach... We caution that future supply constraints could reverse the effect."
I've only heard about datacenter protests in the past 2 years. So perhaps what the authors warned about is happening now?
Data centers directly led to my rates increasing over the last ~6 months where several came online locally without supply increasing. Natural gas prices also impacted it, but the data center buildout over the last year has not helped.
That article is two years out of date, and looking at nationwide electricity usage doesn't account for the local impacts of these data centers.
(* nb4 "it's summer": talking about price per kWhr)
> I'm sure other industrial uses can cause temperature increases too
If anything about this paper is accurate, which it clearly is not, then the >8000MW rejected to the atmosphere from the nearby Palo Verde power station should be raising the air temperature at least 250C, 1km downwind. Which it obviously does not.
Interesting, as we increase our total global power consumption, losses in the form of heat will increase proportionally.
This is more of a zoning problem
"...downwind air temperature warming as high as 2.2 °C, with average downwind air temperatures 0.7–0.9 °C warmer than corresponding upwind areas. Thermal signatures were detectable at distances up to 500 m from facility perimeters"
"The attribution of this warming to data center waste heat is supported by the consistent alignment of the temperature signal with the prevailing wind direction across multiple sites, dates, and meteorological conditions. "
Well, no? That could also mean that the heat island effect of the (statutorily mandated) employee parking lot is causal.
You can look up the NTT PH1 Datacenter from the study on Google Maps. It doesn't have much of a parking lot but due to poor industrial zoning, it's right next to a residential suburb. It's basically a failure of urban planning.
It's in the middle of a large, disturbed area with big commercial buildings, roads, and parking lots, only some of which are data centers. This type of disturbance tends to get hot in the sun. But the larger question is how they are attributing the temperature at the end of the cul de sac to the data center when the somewhat more obvious reason for a hot spot there is that nothing shades it. That is, the measured cooler streets are shaded by the houses, but the edges of the neighborhood are shaded by nothing.
Anyway the main problem with this paper is it does not have an explanation for why the response isn't linear. If areal density of heat rejection was able to raise air temperatures this amount, then nobody would be able to go anywhere near a thermal power station without being vaporized.
All I can think of when reading this was Hank Hill yelling at his neighbor: "Dale you giblet-head, we live in Texas, where it's already 110 in the summer. And if it gets one degree hotter I'm gonna kick your ass!" [1]
[1] https://www.imdb.com/title/tt0620288/characters/nm0362548/?i...
The key is this says at distances up to 500m downwind. This is a nothing burger sorry
Cool, I'll be putting big data ai machine learning data center in your back yard.
No one has ever built a full datacenter in someones backyard...
The referenced 169 MW location is literally across the street from a residential neighborhood, where the article indicates they measured temperature differences[0].
> This gap is significant because many facilities are sited adjacent to residential neighborhoods (Fig. 1), and their air-cooled condenser arrays discharge air at temperatures 8–14 °C above ambient—often exceeding 50 °C during Phoenix summers—with air velocities of 2–4 m/s [15,16], creating thermal plumes that are advected downwind over inhabited areas. For example, the Iron Mountain Data Center (Fig. 1(c)) has cooling equipment located less than 50 m from the nearest three-story apartment building.
[0] https://www.openstreetmap.org/#map=17/33.267398/-111.880313
gives a whole new meaning to 'vertical integration'
will they give me free tokens
Reminder: 4 degrees = 0.0698131701 rad
(article says 2.2 [C/K], btw)
Now compare to how much an asphalt parking lot raises temperatures.
so now we are speed-running global warming
there are now record setting gas turbine sales instead of solar/nuclear
* https://www.youtube.com/watch?v=gk8ziQCWK1w
and for what?
so we can have totally useless "AI" customer service agents where you still have to demand to talk to a human to properly get a problem solved
You haven't used the latest models! AGI is just around the corner! China! Russia!
Nat gas is the cleanest burning fuel we have
That doesn't make the slightest difference for two reasons:
1. Natural gas still produces CO2, which is a climate change driver. (And, since burning more of it will only increase global temperatures, the data center will have more trouble losing its excess heat.
2. Even if a data centre uses a totally clean energy source, it's still excess usage. That clean energy could have been used elsewhere. Until all energy is clean, usage by unnecessary sources is a net negative.
You breathing creates CO2, but we can assume you create enough good for the world to offset that so its okay.
how about the non-stop methane burnoff for months ("flare") from all the oil fracking fields in the US that can be seen from SPACE for the past DECADE
* https://www.washingtonpost.com/news/wonk/wp/2014/10/20/ameri...
* https://science.nasa.gov/earth/earth-observatory/gas-drillin...
that's already raising the local area temps, why not put that wasted fuel to work instead?
build the datacenters literally next to the fracking fields, absolutely no-one there to complain, they all moved away already from the toxic air/water and there's already unlimited fuel
You described the company Crusoe. They've been building datacenters that do exactly what you said.
https://www.crusoe.ai/resources/blog/understanding-the-probl...
> Crusoe has built their business around turning flaring from an operational and regulatory risk for oil and gas companies, as well as an existential risk for our planet, into a multi-solve solution. Crusoe doesn’t just mitigate flaring emissions. They capture the gas, convert it into energy, and use it to power computing infrastructure.
wow Crusoe deserves more press if everything they are saying checks out
flaring has upset me for a decade, even if it was just used for bitcoin mining it would have at least done something vaguely beneficial compared to just raising earth temperature
but if it can help solve the "AI" energy demand boom that would be huge
because it's not like this country is ever going to stop fracking at $80-$100 a barrel now
Methane has a stronger greenhouse effect than CO2 (until it decays into CO2) so it's probably better to burn it and do no work than just vent it, and better still to leave it in the ground.
But ... but ... but ... ai companies have safety teams that ... well ... make sure some kind of singularity mumbo jumbo threat wont happen. The thing they dont care about are ... well ... actual people and actual environment.