1786183626 | New Amazon Data Center Is Set to Have the Most Polluting Power Plant in the U.S. | https://www.nytimes.com/2026/08/08/climate/amazon-data-cente... | https://news.ycombinator.com/item?id=49220350
1786210072 | Amazon Is Creating the Biggest Pollution Source in the Country | https://newrepublic.com/post/214111/amazon-data-center-bigge... | https://news.ycombinator.com/item?id=49223845
1786350058 | Amazon plans gigantic gas power plant for new AI data center | https://www.heise.de/en/news/Amazon-plans-gigantic-gas-power... | https://news.ycombinator.com/item?id=49240853
Sadly there aren't much support to do this. The largest coalition see thermal power stations as critical to manage grid stability and voted to have natural gas defined as "green", since they view it as necessary to enable renewable energy to be built. The other major coalition do want to decrease the need for thermal power plants by increasing base load through nuclear power, but they are not openly against thermal power plants as a way to reduce costs. EU lacks any movement that are willing to pay the cost of not burning fuel in order to generate electricity. Same for basically every other place on earth right now, despite EU and most countries being in agreement that climate change is real and fossil fuels should be stopped yesterday.
Even with today’s energy polices, electricity costs have skyrocketed (this was the case before the data center demand). Couple that with electrification of vehicles, phase out of natural gas/oil heaters, I think its reckless abandon to push policies which harm middle and lower class families. Not to say some of these countries even consider nuclear energy “evil”. Absolutely ridiculous.
The national security nightmare is the dependency on fossil fuel sources. If we were to lose our crude oil supplies because of a war in the Middle East, blocking a strait for tankers, we would be well and truly fucked. We don't produce enough domestic natural gas to replace it.
The only options we have today are solar, wind, and, given enough time, nuclear. These three sources remove external dependencies for generating energy. I'm not sure how much uranium we have in the States, but once the blades are spinning or the sun is shining, we keep getting energy, and no war overseas can stop that.
Electric prices are climbing because we're doing externalities wrong. For example, utility charges for data centers should be significantly higher than what we charge residential and non-DC businesses. The goal is for data centers to absorb the cost of the impact they have on the grid and water.
They could have on-prem power plants, but those plants should belong to the local utility and charge the data centers the data center rate. It also should be a surcharge for the air pollution those fossil fuel monstrosities create. Again, paying for the externalities.
They could get a break if they deployed gigawatts of solar and batteries or used their waste heat for district heating, but yeah, I'm tired of letting them get away with dumping externalities on us. We now have a nicely delineated scenario where we can focus it on them and make them pay.
The more nuanced view is that I would support a residual use of fossil fuels for niche applications that require the properties of the fuel in instances where no known substitute solution would do the job.
By and large, renewables are marching towards a state where they will provide enough of the generation that what remains would be insignificant.
I would be interested to know where the gas is coming from for this particular data centre. There are a few cases that generation by gas is preferable to what is happening right now. One good example is if the gas is from sources that would otherwise be flared off. I have heard in the past that businesses have been reluctant to harness energy from gas rather than flaring it off because any user of the energy gets tagged with the CO2 emissions. If it just gets burnt off it makes no difference to the amount of total emissions, but it avoids the CO2 being attributed to their industry.
Now you will have the AI lobby pushing against EVs (so that oil drilling can create gas as a byproduct), as if it wasn't already an uphill battle in the US...
It's not hypocrisy to want things to be better and tolerate that some things are net negative. I have a fireplace and enjoy using it. But it's bad for my kids health and neighborhood air quality. So I use it far less than my parents did theirs. I also don't commute and fly far less than my father did throughout his life.
Life isn't some strict binary, all or nothing affair.
You've spelt "building' wrong twice, the exact same way 'biulding'.
Very whataboutism
(I think the way through this mess is huge investment in renewables, and that blue states have calamitously dropped the ball by pushing so much data center construction to red states that aren't going to incentivize that.)
This statement can be applied to literally any topic and it fills me with a dread that is only matched by trying to fathom the inifinite nature of the universe and time. The notion of having good faith, well intentioned debate and discourse around any conceivable subject just feels impossible most of the time.
Texas is the largest producer of renewable energy in the country.
And in any case, that lowers Texan per capita consumption even more given that they're not actually consuming all of it.
https://news.gallup.com/poll/709772/americans-oppose-data-ce...
https://www.reuters.com/world/us/americans-wary-ai-driven-da...
People don't just hate the idea of data centers "in their backyard" the way they do with housing, they also just fundamentally despise the tech companies and personalities who are creating these data centers.
The underlying principal components of these data center polls are very probably just:
(1) Concern about electrical bills, which are often valid, and noise, which sometimes can be as well
(2) Political polarization (Democrats --- I'm a lifelong Democrat --- are currently getting marching orders to oppose data centers)
(3) Bogus concerns about water.
We spend a great deal more time and energy thinking about DCs on HN than the median poll respondant ever will. And still, the idea that there's striking public opposition to data centers anywhere in the country is hard to support with evidence.
Fortunately, I think the public opinion fever about data centers is starting to break. I'm reading more and more left-leaning stories questioning the narrative. I think there's a good chance this sentiment crystalizes to something more specific and productive, like "stop letting Elon Musk do things".
Every corporation and even most small businesses have had a machine room, wiring closet, or some kind of server shoved into a room, since 30 years ago now. They all had them in every office building you ever drove by, or lived next to, or shopped near. It's simply that people didn't even register them, because none of the general public really sees or knows what goes on in office parks, and behind the tinted corporate windows. It could be as benign as a toy distributor or as wicked as a defense contractor, and from the outside, they look alike.
But they've all had computers, and networks, for decades. And what we are seeing now is consolidation and concentration. The offices and parks and buildings are hollowed out -- they've laid off workers, they've gone WFH, and most importantly, they've gutted their machine rooms and removed on-prem equipment in migration to the cloud. So each and every company that migrated to the cloud has shrunk their own machine room's footprint by X amount that is reflected in data-center capacity by an identical (or perhaps larger) X. And that, of course, means that power, cooling, and resource consumption is not necessarily enlarged, but merely transferred and concentrated into DCs rather than power-hungry machine rooms, humming secretly inside every office building.
I look back to my first job--I worked at an ISP, and we colocated with the San Diego Supercomputer Center. The SDSC was a data center par excellence with the finest machines that ever existed on the floor. And I was still a kid when I'd be admitted to the main floor, I'd walk past a Y-MP, a Thinking Machine, an Intel Paragon, a Nearline Tape Robot, and there were our humble little pizza-boxes, sitting on a table trying to keep up with Usenet. Our own ISP offices didn't even have a machine room of our own, just a rack with a hub--yeah a hub--and a Sparc 2 pizza box to serve the office. We were already "in the cloud", as it were. More or less literally--SDSC was where they tested nuclear weapons, predicted the weather, and explored outer space, through the magic of FORTRAN...
So what people are complaining about is something they can see. If you're dying of cancer for 20 years but you never go to a doctor, you won't feel it. Once you see the doctor, the chemotherapy starts, and then you croak from the chemo. So this building of datacenters is simply bringing this into the open, and into the light. And people are angrily reactive at social media, and techbros, and surveillance capitalism, and AI-job-losses, and corporate hijinks in general, so data centers do give a focus to that futile rage.
But it's futile, and we all depend on data centers already, even before they are built, and we need DCs now to sustain our economy and our civilization as it already is. It's futile to resist and if you want to come in with a tractor to demolish them, you'll be arrested, and the Flock camera attackers will be hauled into prison too, and technology will just impassively march along as a juggernaut with no feelings, no remorse, and no mercy.
It's a pity; we were really nice folks for a few hundred years.
Success and Scale Bring Broad Responsibility
We started in a garage, but we’re not there anymore. We are big, we impact the world, and we are far from perfect. We must be humble and thoughtful about even the secondary effects of our actions. Our local communities, planet, and future generations need us to be better every day. We must begin each day with a determination to make better, do better, and be better for our customers, our employees, our partners, and the world at large. And we must end every day knowing we can do even more tomorrow. Leaders create more than they consume and always leave things better than how they found them.
these actions are exactly the opposite of what they claim to try to strive towards.
At the current rate it will stop in four to ten centuries.
https://ourworldindata.org/grapher/installed-solar-pv-capaci...
That said there has never been argument about whether we are on that trajectory, we want to be on that trajectory
Or maybe not. Maybe this is a civilizational growing pain these aliens are familiar with. Maybe they went through the same period themselves.
Or maybe you'd prefer this?
So China and India can use all the fossil fuels they want, and other counties too not just them.
But the West has to kowtow to Net Zero / ”Renewables” nonsense?
We’re not buying it anymore.
Everything has tradeoffs.
Destroying the economy and letting low income and poor people freeze in the dark or melt in the sun is passé.
The #1 increase in world electricity demand is people rising out of poverty. China has been producing the most solar in the world and still hasn't been able to cover their own demand increase with it. Eventually that form of growth will slow down. Sustainability should always look at the second derivative.
>GW Ranch received a permit from the state of Texas allowing the gas power plant to emit 33 million tons of carbon dioxide into the atmosphere. If the project emitted that much CO2, it would be the largest single source of pollution in the United States
Immediately following that is the text
"It’s worth noting that companies rarely emit as much as their permits allow"
It's also worth noting that the headline turns "largest single source" of CO2 to "top source" of climate pollution. They say pollution in the text, but context would suggest they are talking about CO2. To say otherwise would be a much stronger claim which I don't think you can assume unless they clearly assert it.
I'm not sure if it is possible to quantify scale of climate impact between different types of pollution, but it seems like a stretch to unilaterally declare that the single top producer of CO2 is the top environmental impact. It may be, but they are making claims beyond what were made in their sources.
I was under the impression that there were a few chemicals that had an outsize effect on climate pollution.
For starters, why focus on the plant being built when it has been permitted. If you focused on the permitting process presumably you could prevent many instances simultaneously.
Edit: https://cleanview.co/reports/behind-the-meter-data-centers
Texas is just ripping through these things since it’s so permissive. That’s troubling for the other states who will all inevitably bear some burden.
Because without context you have no idea what that actually means.
If you get attacked by an unknown number of duck sized horses, how much of an effect would defeating the largest of them have?
How much is the plant actually going to emit? Is the number the largest because the field is dominated by many many smaller things that cumulatively sum to a figure so large that if any one value (even if the largest) is removed, it makes no meaningful difference.
What happens to the fuel if they don't use it? Does it get used by someone else who would otherwise had used renewables.
I don't think it is unreasonable to have the relevant information to hand before making a decision. I might make the same decision as you have already done. I'd just like it to be based on actual information that I have.
Couldn't you just say this about all the others too, then why do anything at all? If you wanted to actually reduce pollution, doesn't it make sense that everyone focuses on a few first, then hopefully maybe the rest could fall in line?
I tried to point out that we're only talking about this because the public ledger makes it easy to measure such things, and we have nothing to really compare it to and we judge no other industry this way.
People raged at me for saying this and assumed I was a crypto shill, in spite of me loathing crypto and never owning any.
All this to say I appreciate your comment and the desire for precise criticism. I hate aggrandizement whether I agree with the position or not.
I think of this as the Wikipedia effect. People perceive Wikipedia as unreliable because they can perceive the mechanism by which it can be wrong. Study after study has shown Wikipedia to be more reliable than other encyclopedic sources that simply assert their accuracy while leaving the mechanisms by which they work hidden.
It is similar to the idea that something that saves 5 unidentified lives at the cost of one identified life is considered far worse than something that costs 5 unidentified lives to save one identified life.
Crypto and AI have also the problem that people cannot understand scale. How much is a million litres of water in terms of swimming pools, or showers? Thing uses as much electricity as [country that uses gas for heating and does not have typically have air conditioning]. Or Training an LLM produces a big looking number of CO2 which is as much as n cars in their total lifespan which makes it look like a different scale. Weirdly, the last figure I read on that suggested that if 10% of Americans trained their own personal leading edge LLM from scratch, cars would still emit more CO2. I don't know if that is true or not, but it frames the entire thing in a different light.
Per kWh, natural gas plants emit half as much CO2 as coal plants and almost zero sulfur and particulate pollution.
They emit little nitrous oxide and almost zero mercury/heavy-metals.
https://www.world-nuclear-news.org/articles/amazon-updates-s...
However, building new nuclear reactors is a slow process. These reactors aren't expected to be built before the 2030s.
China is currently the fastest at bringing new reactors online. Even there it takes 6-7 years from construction start to commercial power generation, e.g. https://www.world-nuclear-news.org/articles/chinese-reactors...
Personally, I think we should look at a structure similar to the military for running the plants. No profit motive, enlistment terms, court martial, stigma associated with dishonourable discharge, veterans benefits, etc. The military is the only one I know of with a track record in nuclear energy that I would trust.
All jokes aside, yes, we should invest massively in nuclear.
SMRs are vapourware and are similarly economically uncompetitive.
Just build solar, wind, and batteries, and we'll worry about solving for the last 15% of electricity demand when that is actually a live problem (for which there are solutions that don't involve nuclear power and don't rely on magic beans being invented).
No need to worry though, on the current trajectory I doubt our civilization will have the opportunity to deal with it.
1. Any equipment or substance that is both nuclear and potentially dangerous has some sort of magical mutant super-power to stimulate superhumanly idiotic behavior in a non-trivial fraction of the people handling/operating/managing the equipment or substance. That behavior is almost always coupled with pathological denials - both from the mutant super-idiots and their supposedly non-mutant peers.
2. General population humans are quite aware of problem #1, hence extremely distrustful when any human experts or organizations try to reassure them as to the safety of anything nuclear.
3. That "reprocessing" ain't nothing resembling "put on a lead apron and pour it through a coffee filter". Instead, it's a quagmire of chemicals and processes that'd still be horrific if nothing nuclear was involved, plus weapon proliferation overtones, plus all the problems noted in 1. and 2. above.
1. Plant construction is very expensive, and not in small part because people who don't actually understand the technology take every opportunity to tie up construction in the courts and regulatory agencies.
2. General population of humans are quite aware of #1, so even though they know nuclear wins on a technology basis, they know it will be unduly expensive, preferring to knuckle under to the mob instead of doing the right thing.
3. People not familiar with the technology don't realize that while we don't generally do reprocessing in the west, it's primarily due to cost, and that Japan will start reprocessing 800 tonnes of fuel per year starting next year.
The technology is the safest way we have to generate power, and could be the lowest cost if the regulatory environment was adjusted properly. It's just a shame there are so many people who are irrationally opposed.
Vs. those with both the unquenchable zeal and bottomless resources to win a multi-decade crusade to prove that they're Ultimately Right (but no short-term need for electrical power) should embark on an epic quest...
Not considering this tradeoff has led to much carbon waste.
I am hugely skeptical that it will be solved economically given that it suffers from many of the problems currently pricing coal and nuclear out of the market.
the politics of the threat of proliferation kills it in the crib.
Solar generation buildout in ERCOT is also proceeding very quickly, the new Tesla megapack factory outside Houston and proposed new solar factory will help speed up the transition to sustainable energy.
Forget newspeak, get out of your bubble and find actual facts from contrasting sources, make your own informed conclusion. Fact finding is not a dirty word. I mean, two words.
And if you do accept any set of those things, When they talk about the top one , there is an assumption that there is a known way to rank all of those things. Show me that before you make a claim that one outranks another, otherwise we are talking about an imaginary scale.
Though it’s so much more complex than that so maybe we form an intergovernmental panel to try to sort this out.
Sure, climate change is bad and has the potential to disrupt and end the lives of many in the future. But other forms of air pollution like pm2.5 particulates in the air are already causing an estimated 5 million premature deaths per year as well as a slew of other health problems including lowering of IQ.
Any amount of focus would be "so much".
I gather you believe other people pay too much attention to CO2.
That the planet is warming suggests to me they do not pay enough.
Why would you think that? I care a great deal about the environment.. I've been working on several reforestation projects for the last 5 years.
I wish there was 100x more funding for CO2 emission reduction. But it frustrates me far more how little attention pm2.5 pollution gets. I live in Asia and see the effects that pm2.5 pollution has on 100's of millions of people every day.