Those in power seem to think that a compute advantage is necessary to maintain economic advantage. Despite this, the voters universally hate data centers, so much so that both NY and TX have data center construction moratoriums.
Despite this, those in power refuse any environmentally aware energy policy nor water policy. Despite this, companies have been trying to be environmentally aware for either moral or marketing purposes to get social permission to build DCs. Despite this, companies are “forced” to build their own “dirty” power sources since political power is disrupting green energy build out. Despite this, demand is still high so that compute costs are continuing to rise. Despite this, our economy is increasingly predicated on a few companies using more compute than geopolitical rivals with cheaper electricity. Despite this, those in power are, again, unwilling to create a compromising policy (aka clean energy) that voters actually would demand before tolerating more DCs.
What’s going to lose first? Voters/environment, politicians/the oil lobby, big tech, or the overall american economy?
Look to Australia if you want to see how these energy policies work out in the long run.
Energy is big money. Anything that disrupts the status quo is immediately vilified. It got to the point that our major party at the time lead a “gas fired” green energy policy.
They got voted out, the new guys implemented a subsidised home battery policy. In the space of a year it has destroyed the economics of a carbon fuelled grid. All the economics pointing to renewables being the way forward were proven correct.
Everyone gets their back scratched. If you’re a consumer and you invested, you have a massively discounted personal power station that generates you money. If you operate the grid, your transmission costs and peak transmission costs went down. General consumer prices went down because of a reduction in peak electricity use driving up retail costs. Everyone in won.
Except big gas, who cannot turn a profit on a renewable success story. Don’t let rhetoric become your reality.
It's absurd how the propaganda against green energy on certain political spectrums essentially turned into, "It's all money/cash grab for green energy companies!"
As if the established industry has no incentive in the opposite direction? Everyone is going to act in their best interests in this system, that's how it's designed, the whole point of government is to mediate the system in the best interests of the whole.
RE "...how the propaganda against green energy...."
One problem is statement are made , that depend on certain definitions , where the definition is not what a reasonable person would expect.
Such as statements about costs of renewables. How exactly is the cost of renewables measured.?
I'd have to say there are lots people who do not believe the government is acting in their best interests. Its one of the reasons why Pauline Hanson's "One Nation" party has gained a lot of traction.
> Except big gas, who cannot turn a profit on a renewable success story.
If only.
Chevron’s Gorgon project will be ongoing for decades and making money hand over fist flogging LNG out of Australia to overseas customers.
It'll be a massive source of previously sequestered CO2 and claimed to be "the largest CO2 sequestration project on the planet* .. with the minor catch of (re-) stashing away a very very tiny portion of the CO2 released. Something, something, not let it out in the first place?
Still, could be worse, at least they're not planning on fracking the Kimberley.
Well the tax payer, would be paying the 30% subsidy for Home Battery deployments, which I understand is severeral $Billion 8.5 AU at present.
One group of loosers will be people who can not afford to participate in this program.
It would seem another group would be users who cannot shift their electrity demands from peak times when electricty will become quite expensive.
The comment is factually correct, It's a inconvenient truth.
Battery deployment is not free.Someone (the taxpayer) pays.
I suspect after they are quite widely deployed. Government could change conditions in future that they must be available as virtual power plant (VPP)
There has been a rush on these since people want access to cheaper power and they believe to avoid the power increases to come. That there will be big increases in power , especially during peak time is the forced introduction of smart meters which allow easy tariff change etc ....
> Power prices to fall for most customers, with bigger drops for businesses
> What is driving [lower prices] is a reduction in the cost of producing electricity. We've seen a lot more batteries and solar systems come into the electricity market in the last 12 months. They've been making the market much less volatile," Ms Savage said.
> "We've not needed as much gas and hydro generation in the evening peaks, and that's what's really cut that cost of wholesale generation."
Yep, exactly. Due to the deranged Net Zero policies of successive Australian federal and state governments, Australia have experienced skyrocketing electricity prices, so any small decrease in residential and industrial prices now is effectively a shit sandwich.
The overall American economy is going to pay the price in the long term. And current foreign policy is reducing trust in America as a reliable and stable partner, which will make things even worse.
>Using the age profile of a country with over a billion people to dismiss its geopolitical power has no credibility.
The first two charts are for GDP, which correlates pretty well with geopolitical power. Moreover if we're going by population, wouldn't that imply India, and then later Africa are going to be superpowers?
A "in general" correlates with geopolitical power doesn't necessary mean you can point at a specific case look at their GDP and come to a good conclusion. Especially not if it's as much of a special case as China.
> wouldn't that imply India, and then later Africa are going to be superpowers?
which might actually happen. If the US falls further from grace and the situations stays like that for a few decades, or the US and another huge power (e.g. China) fall into some costly conflict then India or a union of African countries have a serious chance to become a world super power.
---
=> But more relevantly this discussion is going into red hering argument direction:
- No one argued that you should go _only_ by population, the argument is, that if a country with a gigantic population shrinks hugely they still have enough people left to wage large scale wars.
- Similar no one (should) argue that GDP is by itself meaningful for such an assessment, it's one imperfect not supper reliable statistical point. Especially given how things move wrt. dark factories (i.e. fully automated => no light one as no human there), AI and similar. Yes, if GDP is supper low or high that says a lot, but a lot is not everything and a average GDP for decade long trends in a word looking to face a major time of change means not much in any direction.
- Similar we are speaking about decade long developments, current military strength ist also just another non fully reliable indicator. E.g. Germany post WW1 had close to actual 0/no military strength, and then there was WW2 just a few years later...
So going by one statistic (GDP) is bad, but going by another (population) gets a pass because... China's "special"? I'm sorry, but that just sounds like a special pleading. I don't like this because it just turns arguments into a fact-free zone where objective facts don't matter and the only thing that does is handwavy vibes from either side. Is China going to be the dominant superpower? The supporters can handwave endlessly about robot armies and the superiority of an authoritarian political system. The detractors can handwave endlessly about the demographic crisis and the fragility of an authoritarian political system.
I litterally wrote a comment saying going by a single statistic is not sufficient.
and it's not China being hadwaving "special" but the measurement you are applying really not working well for a country which is only half capitalist and half planed economy. Which is a separate argument from dark factories which is about (non low end) production in China getting increasingly decoupled from human worker and in turn is way less affected from a population collapse then it would be in other countries. This has nothing to do with robot armies, an argument you just made up, even if China would lose half their population they still had more people and in turn can theoretically man a larger army _even if they have a by % much larger population collapse_, they don't need any Robot army. Similar no one said their system is superior. Or do you mean the US, as a two party system with an extremely powerful president is something some would call semi authoritarian. Either way China having done some pretty effective decisions in recent years in some areas doesn't in any way mean it's the better system. Like a Monarch can do grate decisions, but Monarchy is still is a very bad system.
Either way I also can't follow you how you jump from a listing of rough facts to there supposedly being a fact free zone.
What there isn't is a single simpel conclusion.
I mean it's a discussion where a single bad decision of a larger power can easily cause a unpredictable chain reaction which completely changes what are and aren't likely outcomes.
The world is inherently complex and not everything has a simple black/white conclusion, actually most things don't have one. You can only look at all available metrics and do a guess what the future will bring.
You don’t need a numerical advantage to dominate if you have a technological / organizational advantage. The British empire was outnumbered 10 to 1 and still managed to control a large portion of the world for a while.
US government is working overtime to make sure legacy technologies stay around while the rest of the world is investing into the future.
> The British empire was outnumbered 10 to 1 and still managed to control a large portion of the world for a while
It also had the support of a lot of the population of the areas it controlled and at least toleration by the majority. Most of its colonial civil service and military were people fro the colonies.
>US government is working overtime to make sure legacy technologies stay around while the rest of the world is investing into the future.
"Rest of the world"? Europe isn't even a player in the AI race. China might be better placed, but they also lag the US in terms of model development and capital deployed. They've been catching up, but that's not the same as "[American will] be in a similar place to Canada at best". That would require a lot of things to go right for China and/or a lot of things to go wrong for the US.
AI is purely a virtual world play. It may (or may not) make it easier to write code, reply to emails or Slack messages or create PowerPoint presentations. In the real world what matters are energy, natural resources and making (manufacturing) things.
AI will not make it easier to build a house, to harvest a corn field, to assemble a car. AI will not improve weather forecasting accuracy- this is already pretty much 99.999% solved. So AI will not improve standards of living.
All this to conclude that China’s standards of living will improve while America’s will stagnate (or worse)
But sure, go ahead and enjoy those AI videos-some of them are even funny.
Yes, all the technologies of the future demand a huge increase in energy generation that fossil fuels can’t satisfy. In the long run (decades) the US won’t be able to stay in the game with this kind of anti-scientific policies.
I'm not sure if I'm mis-reading your comment, but:
The idea that you "need a single extremely powerful country to replace it or the US will stay in its position" just doesn't add up. I'm not fully sure if that is what you meant but I have seen that argument a few times.
_There doesn't need to be a replacement._
For most of the worlds history there wasn't a single leading world supper power, just regional super powers, or supper powers in specific sectors.
And if we speak about who could become a supper power _after the US stops being one and there isn't one for some decades_. Then there are a bunch of candidates.
E.g. China has a population collapse issues, but due to it's cut throat ruthless government and total population number and lead in factory automation might be from all countries with similar problems by margin the most likely to overcome it without power collapse.
E.g. the EU might have many issues (as it's members has many issues) but the general concept has proven hugely successful. And what people forget is EU is the cherry on top of a huge stack of treaties which can be extended to 3rd parties independent of full EU membership. Like e.g. Canada, which doesn't plan to become an EU member, but might join some of this treaties as result of the non stop hostility from the US. Same for military treaties between NATO members which are independent of NATO like e.g. JEF, which would likely be open for letting Canada join. Anyway way to many words. But the end result may be a world supper power which is not one entity but a complex lossy coupled network of countries.
Then there have been some other cross country unions and countries, which want to gain super power relevance but have so far been too irrelevant due to a combination of corruption, bad luck and sabotage by existing powers etc. But if we speak about going for a few decades without any leading super power and maybe prime canidates harming each other (e.g. a US <-> Chineese war) then they do actually have a chance.
>The idea that you "need a single extremely powerful country to replace it or the US will stay in its position" just doesn't add up. I'm not fully sure if that is what you meant but I have seen that argument a few times.
The OP goes further than merely claiming that US will lose its spot as the top power. It claims it won't be a world power at all:
>in 50 years America won't be a world power any more. It'll be in a similar place to Canada at best
For reference, Canada ranks #11 in GDP, below Russia and even Brazil. On military spending it ranks #16, below Poland and Spain. It doesn't have nukes. To suggest US will ever come close to that in 50 years is laughable.
50 years once just a blink in human history is today enough to have 2 major industrial revolutions
nukes are not an argument for anything but main land defense from invasions
Look Russia, it was considered on of the most powerful countries military wise and cleanly in the top 10 economically. Many treated it as a 3rd place behind USA/China. It probably will be neither military nor economically overly relevant in 10 years. Through comparing a war economy with peace economies by most normal usable standards (e.g. GDP) isn't very meaningful. Yes it has nukes, no one will invade it (in a traditional large scale sense out side of some border conflict). But it's economy looks like the kind of war economy which leads to a complete economical collapse once the war ends (i.e. burning core substance) end while it's military could be very experienced now, by treating soldiers as highly disposable its unclear how much that actually is true.
Now look at the US, it has lost the majority of soft power thanks to incompetence (and it was maybe the nation with the most effective soft power). It has alienated most of it's allies, especially it's closest allies. Both from a military and economical POV. It's highly dependent on China, and tries to change that have shown very mid to bad outcomes. On the other hand China is increasingly less dependent on it (and the EU) and increasingly less attackable by sea blockades and other non-hot war counter measures the US had. It's behind on automated production. Countries are increasingly less willing to tolerate US bases on their ground which are essential for it being able to project power. The economy is increasingly dependent on a few pillars, which are strong but can crack anyway. The might claim AI leadership, but practically it looks like the price they payed to get there is one which not only will not pay of but can not pay off, it's just way to high. War is also increasingly going back to ways where even if you have superiority the cost of waging it is still severely high, in such a situation you can't force leadership by military force, if you try it likely will end similar to Germany in WW1. Lastly the internal instability Trump has brought by dismantling checks and balances
and arbitrary gutting of programs which turned out pretty essential for stability is bad enough to argue things like a putsch series or civil war being _realistic possibilities_ (not guaranteed at all, but possible where it wasn't realistically possible before).
If you put everything together the US dropping temporary out of the top 10 of overall most powerful countries is, not very likely but, very much possible.
It does have absurd geographical advantages, and nukes to make sure it keeps them basically for ever (as long as it doesn't splinter which isn't very likely tbh.). So the drop would be most likely be temporary at most.
Still it is possible, it's the price to pay for first squandering all trust/goodwill gained from WW2 over a few decades of greedy, egoistic, egocentric, ethical catastrophic politics. And then going complete "unpredictable, untrustable, instabile, high risk, publicly considering to start war with NATO or at least half of the EU (JEF+some others)" in recent years.
>> nukes are not an argument for anything but main land defense from invasions
But it is not an argument even for that. Imagine Russian tanks are crossing the border towards your capital. Will your government use nukes is such case?
lol… you don’t know anything about the Chinese government if you think they aren’t already preparing for this. And the difference between them and us is that when they prepare for something, they actually prepare for it with real state capacity and planning. You know they are building armies of industrial and humanoid robots for exactly this reason?
Also, sorry but the economist hates China with a passion, you really shouldn’t believe a thing they say on the topic.
>Also, sorry but the economist hates China with a passion, you really shouldn’t believe a thing they say on the topic.
Note I'm just linking charts and figures, not their editorials. It's ironic that you accuse "economist hates China with a passion", but the best refutation you can come up with is some handwaving about future robot armies. How do we know you're not just hating US (or loving China) with a passion?
There's so many ways to slice the data, I'm not sure there's an apples-to-apples comparison between US and China. However, it's fair to say both economies face demographic headwinds.
If Canada was a US state it would rank around 4th or 5th in terms of economic size and power. There's a lot of doomerism, but I seriously doubt the US is going to lose its top 4 or 5 economies in the next 50 years.
If you want to go really doomer, then I think it's actually more likely that Canada would become a US state instead of the US economy shrinking to the size of Canada.
Speaking as a European, current foreign policy is not "reducing trust in America as a reliable and stable partner". It has already reduced trust to rock bottom.
It's hard to see how much worse it could get, short of outright invasion or allying with Russia.
Just this week, amongst other news:
* America's useless adventures in Iran have used up so many Patriot missiles there aren't any left to defend Kyiv against Russian missiles, while Iran is still firmly in control of the Straits of Hormuz.
* Meddling in Greenland went up a notch with landing equipment to start oil drilling.
* State Department funding MAGA-oriented "free speech" disinformation campaigns in the UK.
* The revival of Trump's tariffs under the hypocritical guise of guarding against slave labour. Especially when the US prison system is such a large domestic source of slave labour.
Zero trust left here. Zero desire to visit the US ever again.
I feel it is hard to grasp for an American how much perception of US shifted in eu. This is just anedoctal of course, but a famous Italian singer famous for having written a song about American and sexual liberation decades ago, wrote an updated version. The former is "America", the latter is "America INC." You can guess which one depicts America in a bad light.
I wouldn't have expected a famous singer to make such a song, given that America was always portrayed in italian pop music with a positive image, and negative views were relegated to mostly political groups or singers with political themes
Simply put, I think Americans don't grasp it because even for us Europeans it is sometimes surprising how repentine and fast the shift has been in the regular Joe's brain - I'm not saying that Americans are stupid bla bla thing -, I'm saying that it is normal not be as up to date with some phenomena as people living where that phenomena happens.
Also, at least the Italian government is still dancing the tip tap regarding a position on America, not taking any definitive stance or doing anything significantly different than before, which is a big rift wrt what people want. Only exception for defense expenses
I don't think the first group understands how important Europe is to America's status as superpower, while the second group operates under the delusion that the world will go back to business as usual after the next presidential election.
Only the elderly who are blind to the decline and can’t even picture a world order outside other than Pax Americana believe we will go back to business as usual. The rest of us realize it will take decades of stability to gain a semblance of the trust we used to be granted. And in the mean time the rest of the world will forge ties that don’t include us. We will be weakened on the world stage compared to status quo prior to trump for the rest of my life I’m sure.
There is a lot of that with the population but I don't see it at the leadership level yet. The top politicians are all still desperate to go back to the good old days of euro-atlanticism (and all the cushy positions it gets their friends & family), there's also the problem of all of their capital happily sitting and growing in fortune 500 companies.
> It's hard to see how much worse it could get, short of outright invasion or allying with Russia.
Why would the US ally with Russia, who is a Chinese and Iranian ally?
That's just uncritical thinking.
The United States is the single largest bilateral provider of direct military equipment, intelligence, and defense funding to Ukraine (a European country) against Russia.
> Why would the US ally with Russia, who is a Chinese and Iranian ally?
I don’t think the comment above you is suggesting that is a possibility, but rather saying “things are so bad that the only imaginable way to get worse would be <outlandish unimaginable scenario>”. In other words, it seems to be hyperbole.
Perhaps, though my reading between the lines is that they think trust is so low because the US won't help defend Europe and might invade Europe, even though, despite rhetoric, all actions point to the opposite happening.
It is, as the USA has a large population. We can be more nuanced.
By absolute dollars, single country: US > EU institutions > Germany > UK
By absolute dollars, bloc vs. bloc: EU/Europe combined (~$237bn+) now exceeds the US (~$130bn), especially after 2025
By per-capita/GDP burden: tiny front-line states (Denmark, Estonia, Lithuania, Norway) far outstrip both the US and the larger Western European economies — the US actually ranks closer to the bottom of the pack on a relative-effort basis, behind Germany, the UK, and virtually all Nordic/Baltic states
Between February 2022 and January 2025, average expenditure per European resident stood at $394; by April 2026 that figure had risen to $618, driven by the total spending by European NATO countries on Ukraine exceeding $237 billion, including $96 billion contributed just since January 2025.
Average European (across NATO's European members) $618 per person (2022-2026)
United States (my estimate: ~$130bn total ÷ ~340m population)≈ $384 per person
I admit I haven't followed the French protests as closely as I should have. Do they shoot protestors in the face with teargas canisters there too? Maybe they do. Do those protestors face the real possibility of medical bankruptcy if they need extensive care? Does France enforce laws that protect striking workers?
The problem with the US is not the regime, it's the people. We elected Trump a second time with ~50% of the vote after seeing four years of his complete buffoonery. His approval rating is still around 40%. If he were somehow to run again in 2028 (never mind it being unconstitutional), there's a pretty solid chance he'd win. His party still backs him 100%, and nobody has any idea whether they'll maintain control of Congress after the midterms, because they still have enough support to plausibly keep their majority.
It's not about being too meek to fix things. It's about half the country wanting what's happening.
The environment's already lost. We've passed the 1.5ºC deviation mark from the pre-industrial global average, and we're on track to hit 3ºC in a few more decades.
The excess deaths and human misery (both that which is increasingly clearly ongoing, and we have to look forward to) are going to be tragic.
It banned foreign (edit) made equipment because it’s known for cybersecurity issues. It was the FCC that restricted new foreign-made connected inverters over grid cybersecurity concerns. [1]
> This week, the US government paid a wind farm company $1.2 billion to stop building wind farms.
The $1.2B is real, but it wasn’t payment to stop construction. RWE agreed to surrender three offshore wind leases and settle related claims. Anyways, RWE’s three leases were in very early development and weren’t expected online until the 2030s. [2]
You can still think both policies are bad. But those are different claims.
That’s true. China dominates globally but I found that the US installed European and Japanese models. There is a big supply chain concern and set of reviews happening all across the entire government sector. Energy is a national security concern. [1]
Your other comment got downvoted to death (not by me) so I can't reply to it, so I'm replying here. Sorry the downvoting happened; your comment wasn't incorrect. You're right that inverters with zero connectivity whatsoever are not banned; my comment was overly broad. That said...
The ban covers any inverter with any wireless communication whatsoever. Bluetooth is specifically called out, despite necessarily being local-only and not enabling an internet connection.
I suppose you could make the argument that with Bluetooth, the app that communicates with it has internet access and thus it's the same thing. However:
The ban does not cover a power inverter with no integrated comms, but has a serial port into which is plugged a "sold separately" wifi dongle. Neither does the ban cover an inverter with an ethernet port. It covers solely wireless communication.
> Even if I assumed maximum malice, how are these inverters going to perform a coordinated attack on the grid?
They’re network-connected and remotely controllable. The concern is compromise of the vendor/cloud infrastructure allowing large numbers of inverters to be manipulated simultaneously.
Or even Trojan horse style malware is a possibility I suppose (Stuxnet).
That’s basically the threat model that they are using. If people want to argue with that, then that’s something to poke at
AC power needs it's sine wave aligned or it's very bad. If a huge chunk of inverters all start dumping power onto the grid that isn't synced, things break.
You can't just "not sync the grid". I think what you mean is if all the country's inverters that are curtailed stop being curtailed and pull the grid out of frequency because they control so much power. Note that every inverter as well as every traditional power plant and many substations will disconnect themselves long before reaching 59 or 61Hz - the most you achieve like that is a blackout.
Modern data centers can be designed with near-zero direct water consumption. It isn’t fundamentally necessary to consume huge amounts of water to cool computers. What’s the story with those new data centers? Do they actually choose to do evaporative cooling that people are complaining about or people are complaining about problem that doesn’t exist?
How? In most locations for much of the year you need to reject heat either by evaporating water or using electricity (more water) to run a refrigeration cycle.
It doesn’t matter what the actual reality is. People think that data centers waste water, and they don’t like that.
The beef from a single hamburger uses far more water than a year of ChatGPT usage. Few try to reduce consumption in favor of more efficient foods by comparison. It’s about perception.
Exactly, this is all a symptom of underinvestment in grid capacity, and datacenter companies having to cobble together whatever electricity source they could get their hands on at the last minute.
It's a complete failure of the state in having any kind of foresight.
China, in comparison, has around 2X excess electric production capacity in their grid. That's what made mining Bitcoin, and now ultra-cheap LLM inference, economically viable.
> ... and this American people have gotten what they voted for.
Probably one of the biggest flaws of a Democratic system is that a voting majority can mandate a situation that harms literally everyone against the will of everyone else even when those harms are well documented and absolutely provable.
"The majority voted for it" isn't a great answer for the question "Why exactly are we letting the bus driver drive us all over this cliff edge to our fiery deaths?"
"Stop the bus! I want off!"
"Nope. The vote's done. You're stuck with the result."
"Well, what if I voted against driving over the cliff edge?"
Bear in mind the US electoral system is far from the best that can done.
Not all democracies are equal and the US system is barely "democratic" - it's spiralled in a two party not broadly representative Hotelling's Law swampland.
> Those in power seem to think that a compute advantage is necessary to maintain economic advantage. Despite this, the voters universally hate data centers, so much so that both NY and TX have data center construction moratoriums.
I don't understand what "despite this" is doing here. People have thoughts about what good/bad policy should be, independent of what good/bad policy should do all the time.
This is not an argument for or against data centers, but rather against populism. People have too many feeling about too many things, that should simply should not be decided by feel (and, sometimes, not enough about things that should).
The sense of urgency is strange and feels extremely forced. There’s no way this is all so important that all other concerns automatically go out the window so projects can complete construction yesterday.
It really wouldn’t hurt if the developers involved tried even a little bit to be good neighbors and to select sites where they’re not going to be a nuisance and net negative to adjacent communities at minimum.
The blatantly extractive, externalizing mindset of this all reminds me of the attitude that mining companies have taken towards Appalachian states for over a century at this point.
Given local grassroots opposition, the data centers are starting to move into less regulated countries where I would gather building them raises the local standard of living, but presents the new challenge of finding people qualified to run them.
This is just pure histrionics, you have no understanding whatsoever of the scale of data center water consumption, electricity consumption, new capacity coming online, etc.
Not much. 0.04% of US water production. For context, that's about 3% of the water consumption of our golf courses, and less than 1% of what we use to water our lawns.
All US data center water usage is less than the total water used for golf courses or the total water used to grow alfalfa. Less water used in evaporative cooling means more electricity is used for cooling given that the efficiency almost doubles if you add evaporative cooling to a chilled water loop.
It’s bad to pump non-renewable fossil aquifer water in Arizona for evaporative cooling, but using cooling towers in an area that gets 30” or more of precipitation per year is probably not going to hurt anything.
The question is does DC building bring added costs to local communities. My small town has a backlog of home builders waiting for permits because of constrained municipal water capacity. We are still paying off the bonds for the last capacity we added. Golf courses/etc are already factored into current costs. Additional new use brings significant cost (bonds to build new capacity), funded by increasing expenses on the entire community, not funded by those who financially benefit from the new build outs.
The issue isn't comparative usage to existing hookups, it's new cost burden on the local community where a data center is built. Which is why DC propogandists switched to the golf course measurement, it confuses the original concern, the cost to the community of new DC hookups to municipal water or the very limited amount (of expensive to the community) available hookups that were intended/funded by the community for other purposes.
> Despite this, those in power refuse any environmentally aware energy policy nor water policy.
Can you tell me how much water a data center uses? If you have strong opinion on the construction of data centers I hope you can answer this basic question about data centers. (Many opposed to them cannot)
As a comparison, one acre of alfalfa growing year round in California uses about 5*325000=1,625,000 gallons of water a year for four harvests, 4400 gallons of water per day per acre.
There’s around a million acres of alfalfa grown in California for a total water usage of 1,625,000*1,000,000=1.625×10¹² gallons (or 5 million acre-feet) per year.
365 million gallons of water per year for one data center is 1000 acre feet of water. California’s alfalfa crop uses as much water as 5,000 of these data centers.
1000 acre feet is approximately 40 hectare meters. Or, you know, about 1.2 billion liters.
Over on this side of the ocean, we’ve got freshwater lakes larger than entire European countries. There’s a lake over here, it has an island on it. That island has a lake, which has an island in it. And there’s a lake on that island.
We need units of measurement that represent large things. If you don’t, that’s ok.
Do Alfalfa growers traditionally use limited (and funded at large expense by the local community) municipal water hookups to grow their crops? Are alfalfa growers asking for new water hookup permits to access these resources?
Because DCs are, and that is the concern local communities have. Not random comparative statistics whose purpose seems to be to take valid local community concerns and pretend they're just a joke. How much do alfa growers in California add to my water bill? How much will new municipal water hookups by new data centers impact my water bill? One is a valid concern. One is just a BS attempt to discredit the valid concern. Because that is how this shady industry operates. On BS. On manipulation. And they wonder why they are more and more disliked.
It is a sort of "damned if you do, damned if you don't" situation.
The reason these DC use so much water is because it uses less electricity to evaporatively cool than it takes to cool with a heat pump.
You are trading water scarcity for energy scarcity.
In my personal opinion, the right tradeoff is more electricity. Also IMO, these DCs should be required to build out solar + battery + wind offset plants which supply their energy either directly or indirectly by selling to the grid operator at a discounted rate. If you want 40MW of power, you need to deploy 40MW of solar (not peak, but yearly average output) and a 400MWh battery plus all the grid hookups required.
The end result for residence should be lower energy bills, not higher bills.
> Do Alfalfa growers traditionally use limited (and funded at large expense by the local community) municipal water hookups to grow their crops? Are alfalfa growers asking for new water hookup permits to access these resources?
No, you already know they use flood irrigation.
Why is a municipal water system allowing a data center connection that will strain their distribution network? They’re well within their rights to say no, Mark Zuckerberg isn’t holding a gun to their heads.
> Why is a municipal water system allowing a data center connection that will strain their distribution network? They’re well within their rights to say no, Mark Zuckerberg isn’t holding a gun to their heads.
Deep pockets have a way of bypassing practical concerns. [1]
Zuckerberg likely isn't holding a gun to their heads, instead it's more likely he's putting wads of cash in their pockets.
So even when xAI (which is now Tesla? Or SpaceX maybe?) was "operating dozens of unregulated gas turbines that worsen the air quality in one of the most polluted parts of the country"[0] this didn't place them above this new Amazon not-even-built-yet data center? Are they competing who can destroy the earth the fastest or what's going on?
It seems wrong that Amazon should get in trouble for this when it was XAI that began this new race to the bottom.
Peoplecheered xAI on and admired how quickly they were able to move, without much regard to the methods or legality. IMO competitors had no choice but to respond in kind when they see XAI not just getting away with it, but also getting a pass on legal issues.
> It seems wrong that Amazon should get in trouble for this when it was XAI that began this new race to the bottom.
Who gives a shit who started it? The ones who continue engaging in it, should clearly be punished as they're working against humanity's best interest of keeping this planet survivable.
If you're one of the largest companies in the world, constantly brag about how environmental friendly you are and how much you care about "sustainability" or what the trendy word is today, and you end up building a data center that sets a record in pollution (either directly or indirectly), you're doing something massively fucked up and are misaligned with the rest of the people on this planet.
> Peoplecheered xAI on and admired how quickly they were able to move, without much regard to the methods or legality. IMO competitors had no choice but to respond in kind when they see XAI not just getting away with it, but also getting a pass on legal issues.
I guess this is what late-stage capitalism gives you ultimately, companies who feel like it's fine doing whatever because competitors are also doing whatever. No one I know have been cheering for anything like that, I suggest if you have people cheering about stuff like that near you, get far away.
So where is the police? They're running turbines they haven't been granted permits to run, if Average Joe did so the police would show up and literally grab the hardware from him, one way or another, or disable it at least. What the fuck are they waiting for?
One would think if there is even a suspicion (even if it hasn't been ruled in court yet) about large-scale harmful toxic emissions, the police and relevant authorities would at least quickly do some samples or what not, to figure out if they need to put an immediate stop to it until the courts ruled either way.
But seemingly it's still ongoing and no harm no foul?
Do you know the difference between civil law and criminal law? The police don’t care if they’re not violating a criminal statute. If you want to test how it works, build a brand new house in a city without pulling any permits.
I do indeed. Do you know that severe violations of civil law sometimes amount of up to criminal law, especially if there is evidence of criminal intent or "knowing" violations of the law but proceeding anyway?
This is the case for lots of things. Negligence can turn into "reckless disregard for human life", breach of contract could amount to fraud, and many other cases.
Indeed, but they do amount to "suspicion" and are worth investigating instead of saying "Lets wait years until the courts decided if it is or isn't actually polluting" and not shutting it down until they've clearly demonstrated it being safe or not.
All this AI doomerism is hiding the obvious and immediate harm that will cause a mass extinction event. Funnily enough people at Anthropic are predicting the end of the Anthropocene, completely absolving themselves from any responsibility for the actual reality that the Anthropocene is defined by human made geologically noticeable changes. Like causing a global climate catastrophe.
It's possible to have more than one existential risk at a time. We even have a third, since nuclear war doesn't seem completely impossible these days. We probably shouldn't ignore any of them.
An all-out global nuclear war (all nuclear nations landing 10% of their warheads in populated areas) would be a disruption comparable to COVID at best.
Can you elaborate on this? I’m not an expert at all but this seems so far from true I’m struggling to understand if you’re serious.
Wouldn’t we expect tens of millions of casualties, nuclear fallout rendering large swathes of countries uninhabitable, and ensuing total societal collapse once the globalized western world is separated from their foreign means of production?
Of course you're correct. Here's a 2019 simulation of escalating full-scale war between the US and Russia, done at Princeton based on what's publicly known about actual force postures. Immediate effects would be 34 million dead, 91 million total casualties, and 30 major economic centers destroyed on each side.
Fallout would cause significantly more casualties. Then there are the environmental effects. Science on that has evolved over the past few decades, so wikipedia is a good starting point for an overview.
Since none of this is certain and we're talking about risks, considering the worst cases seems appropriate, and those involve agricultural collapse, several billion dead, mass extinction, and an ozone hole covering most of the planet. It might not kill off the human species, but modern civilization would struggle to recover.
It is the paperclip AI scenario:
The AI is cheating humans to produce only one thing - but in reality today, its not paperclibs, but datacenters as "the most important product": So the scenario still holds :-D
The existential risk people are worried about a much shorter timeline - like the next 5-10 years. Climate change is serious but very unlikely to lead to anything like ending the Anthropocene in the next 5 years.
400,000 people in Europe are displaced. Pretending that this is a future issue is tremendously disingenuous. Things don't need to be civilisation-ending to cause a degree of harm that is worth focusing on.
Note that this is about absolute pollution, not relative to power output. Pollution is high, because power output is high. Relatively, for what it's worth, it's just a fairly normal gas plant.
So the most interesting thing to take away from this: There is a lot of demand for more power and in this instance it's satisfied by gas. Which is still much better than coal. Which is still much worse than renewables.
On this map I can't see _any_ nuclear under construction in the US yet the rest of the world seems to be building many.
When a man has a hammer everything is a gas turbine. GEV stock has treated me well but I'd much rather nuclear was winning.
It's a catch22 problem because nuclear is much slower to build than a data center so there is no demand from the data centers. And the nuclear plant doesn't know the demand will be there in 5 years once with gas soaking up demand. So the build it and they will come strategy is risky. Building infrastructure is like a muscle that gets stronger with repetitions. The cost will only come down once some are built but nobody wants to build at the current high prices. Govt help is needed to kick start it.
DOE is investing heavily in SMRs right now. There's 4 which have reached criticality so far. Chris Wright the DOE head is a o&g political appointee hack, but the career employees at DOE are working on pushing clean energy forward.
In 2025 China built 696 TWh of new wind + solar power and 13 TWh of new nuclear power. So to a first approximation it's fair to say that even China isn't really building out nuclear power anymore.
Aren't coal power plants waaaay more polluting than natural gas? Not just in terms of C02, but also really nasty stuff like particulates, sulphur, and metals?
It's very trendy right now to get mad at datacenters, but I'd much rather energy come from natural gas than from an existing coal power plant. (Obviously, renewables would be better than either.)
The nasty stuff, yes. Regarding greenhouse gases, natural gas has turned out to be about as bad as coal once you account for methane leaks, and the industry has strenuously resisted any decent mitigation for that.
Ok. A serious question.
I frequently hear that wind and solar are the cheapest forms of electricity. Then why are they using gas turbines? Is it because they can generate more power with a smaller real estate footprint?
Texas specifically has heavily politicized and demonized wind power, in spite of having massive successes with it prior to 2016 and being a leading producer of it. Solar isn't as bad, but is still seen as "leftist." The political landscape in TX is vastly more friendly to natural gas.
In other words, they are using gas turbines because the politicians are stupid.
They are using gas because it's more reliable during peak usage. Wind generation in Texas isn't optimal throughout the entire state and conditions aren't optimal all the time. They still use wind, but from a pure numbers perspective, gas infra already exists, the gas is there, and gas is more reliable, so wind isn't always the cheapest form of energy. There are also diminishing returns on just adding more wind turbines.
Gas is also incredibly cheap in Texas. Way, way cheaper than you would think. In some parts, it's negative price - you will be paid to use it. We have entirely too much gas, which is actually a problem because ideally we want to discourage its use.
So yes, wind and solar are cheap. Depending on where in Texas you are, gas is cheaper.
Utility scale, solar installs, and will be total solar by end of year.
Texas accounts for 40% of new solar capacity installed per year at this point.
It does lag behind an inefficient and expensive rooftop solar, and has no State mandate for homeowners to build it.
That said, I think this is enough to show that Texas doesn't hate solar, so there's probably a different reason for selecting gas to support the data center. If I were to Hazard I guess, it would be stability and demand curve
Texas is not at all against renewables. Maybe, just maybe, whatever axes the people building data centers are optimizing on rule out solar/wind over factors other than "renewables are for leftists, so fuck that we're gonna burn gas even if it's more expensive."
Did you miss the 1st utility part? It can be quite hard for nonresidential users to retrofit for solar. Not impossible but sometimes not an easy solve.
This is... Not a huge statement to me. My home state is at 16% solar, and we're calling it inadequate. [0] Its not a 1-to-1, as we have probably 10% of the people, and 1/3 of the landmass. More spread out, different demand, so a completely different landscape.
Avoid it to the tune of being in the top third (for now, they're projected to keep growing), behind only blue states that are known for aggressive rollout and/or climates that are overall more favorable (i.e., Texas gets its fair share of sun but not quite like CA and NV, for example)? Is that your idea of avoiding it because politicians want to own the libs?
I often wonder why conservatives have such a problem with renewable energy.
I can certainly believe that oil and gas pay better bribes, and that this incentivises a certain kind of politician who seem to be over-represented among conservatives in the US. But that doesn't explain the widespread dislike among the conservative end of the general population. Is it just groupthink or Overton Window -nudging using social media? Or is there something more psychological going on? Some kind of disgust reaction?
One would think that self-proclaimed "conservatives" would like to conserve the natural world... Long gone are the days of Teddy Roosevelt that's for sure.
Under current administration, I'm not even sure its possible to build out green energy for this sort of thing at scale - this article[1] was on the front page just yesterday
Sure, you don't pay millions of dollars in engineering for a permit you obviously aren't going to be able to get. That's a big part of the problem, they're unpredictable.
Because the US has some policies that make those currently more expensive than they should be while. A combination of tariffs, removal of incentives for local producers, etc. has caused prices to be higher than they could be. All of this is because of fossil fuel lobby funded campaigns for exactly this to happen.
The flip side of course is that gas turbines are in short supply and manufacturers are in no hurry to build out production capacity for what is effectively obsolete technology with a very uncertain future. A lot of the projects being announced may never get delivered for the simple reason that there are long waiting lists for these things.
And there is the notion that related investments might end up under water when the few plants that do get built start struggling economically in ten years or so. These are not very attractive long term investments. The short term interest in securing any capacity at all is overriding this. Of the well over a hundred GW of data centers planned until 2035, probably only less than a 30-40% will actually get built; if not less. Michael Liebreich is predicting only about 35GW will be realized. His estimate is that there simply is no more production capacity that can be committed than that.
It will be interesting to see if this Amazon plant actually gets built. For now Amazon can afford to gamble on this and just write it off if it doesn't work out.
If policy changes in a few years, I expect a lot of these projects to get unceremoniously cancelled when it turns out that they are a combination of under water financially and behind on schedule because of component shortages anyway. Also, it only takes a few failed projects for everyone to get cold feet. Premature closures of new gas plants could happen if renewables get cheap enough. Those are already happening in some places.
For a DC it's probably reliability. You need wind to be spread over a large area to even out random changes in power availability. And, obviously, solar power works during the day.
Renewables require storage because they are bursty - sun goes down, wind stops blowing. This is more expensive than just burning as much nat gas as you need.
Presumably, the data center usage could also be bursty, right?
More people will be using AI during the (work)day, than at midnight. Then they could do half solar, half gas.
Also "more expensive" when it comes to solar isn't meaningful until you also consider timeframe. Solar panels pay themselves off in a few years, and after that they basically generate "free" electricity. Gas never becomes cheaper or free, every year it will cost money.
Next: I even wonder why frontiers labs don't sell batch processing and spot prices to keep 100% utilisation. If they do that and I suspect they will it will be 24/7 maxxed.
Imagine this: Fable at 98% discount because but you need to wait. Then imagine: bots that scan your tickets for easy low priority tasks to do when those cheap tokens come.
Let alone research batch compute needs. Training and so on.
One concept will drive this shift: GAAP profitability.
AWS is heavily used worldwide, not just in the US. The same goes for the big AI companies.
Timeframe is accounted for if you look at levelized cost per kWh. The levelized cost of natural gas is 5 to 11 cents/kWh. In the sunniest spots, the levelized cost of solar with enough battery storage to provide 24/7 power is about the same, if you want >90% reliability. The usual four hours of storage is only enough to account for the "duck curve" during the day.
The cost of energy varies widely based on regional factors, geography, existing energy portfolio makeup, etc. Additionally cost is not just the unit price but the price overtime to support a specific SLA.
Texas has very cheap gas; it’s also good for solar. They are very competitive price wise, so it’s kinda depressing they are using gas. But it’s probably a slight price edge plus better reliability.
Zero data on Greenhouse gas CO2-equivalent per kWh. A combined-cycle gas power plant is going to have lower impact per kWh than the 200+ coal plants in the US.
really time for these companies to have to switch to options like Nuclear SMR from providers like Rolls Royce who already have the megawatt solutions ready in the market, which don't release giant volumes of CO2 which we cannot afford.
Divest all your stocks from all this AI data-center race bullshit, invest into renewable projects. Worst case, the orange buffoon will bail you out with taxpayer money to prevent the US grid from being too clean or something. Only money talks in the US.
A lot of new data centres have been announced in Alberta, Canada. Which is a weird location given that Alberta has a derelict power grid that can barely serve the current load, and runs almost entirely on fossil fuels.
Meta, for instance, plans a $13B data centre in Sturgeon County.
They all plan on powering them by building massive natural gas generating stations with each. In a free for all race to destroy the environment and doom humanity (aside from your plutocrats who'll live in their bunkers). Until a week or so ago Meta still had a clean energy pledge, but finally they just abandoned that silly greenwashing lie: https://finance.yahoo.com/energy/articles/meta-quits-clean-e...
We are basically in the end game now, and despite that weird guy running through these comments telling everyone how great everything is going to be in a hundred years, things are looking utterly grim.
Much more the land of oil and gas. Alberta no longer burns coal for electricity.
The dominant political party frames natural gas as "clean", because it isn't a dirty as coal; and Alberta natural gas in particular as "ethical", because not Russia or Saudi.
I think I get your point; that the inverse of corrupt (I call it fake) capitalism is not necessarily communism, or something reactive. It might just be a system that works better, even if we have not named it yet...
Downvote me all you want. China is doing something right. Nuclear, renewables, whatever the mix, they're not parking gas turbines next to a data center and calling it a power strategy. That US tech companies reached for that first tells you everything. Speed over consequences, every time. And shut up about that China still fires a lot of coal. They are phasing it out... the right way.
You know, we don't actually have to do this. The people could block the gates and prevent the delivery of the turbines. The local fire department could fail their inspection and prevent the gas from being turned on. The UPS could "mysteriously" go up in smoke during it's commissioning.
None of this is preordained. We, the phyical persons, can choose how we want to live, and what constraints the corporate persons must accept.
All this at the very low cost of: your job.
You can't have an unbreachable blockade 24/7 in front of there, very soon they'd send police to disperse. Fire dept has to justify the failing, if everything up to code and the inspector make it fail it will delay for at most a week and the guy will lose his credibility. The guys delivering and installing the UPS are just trying to make a living too, destroying the thing will get them fired while only delaying a little the time getting another delivered. You are replaceable.
Bitcoin received a lot of backslash due to its power usage and ecological footprint. It's obvious that HN users don't have the same criticisms towards AI data centers. Maybe someone with a short position will coordinate that communication soon..
It’s worth noting that the ideological drive to create more energy-neutral crypto schemes resulted in trillions of dollars in fraud and no value created. Same as it always is with climate extremism. Always a cash grab.
What is climate extremism? Because it doesn't seem like the world is actually doing much to combat climate change as we see more fires and more extreme weather while still clinging to the old fossil fuel economy as long as possible.
Those in power seem to think that a compute advantage is necessary to maintain economic advantage. Despite this, the voters universally hate data centers, so much so that both NY and TX have data center construction moratoriums.
Despite this, those in power refuse any environmentally aware energy policy nor water policy. Despite this, companies have been trying to be environmentally aware for either moral or marketing purposes to get social permission to build DCs. Despite this, companies are “forced” to build their own “dirty” power sources since political power is disrupting green energy build out. Despite this, demand is still high so that compute costs are continuing to rise. Despite this, our economy is increasingly predicated on a few companies using more compute than geopolitical rivals with cheaper electricity. Despite this, those in power are, again, unwilling to create a compromising policy (aka clean energy) that voters actually would demand before tolerating more DCs.
What’s going to lose first? Voters/environment, politicians/the oil lobby, big tech, or the overall american economy?
Look to Australia if you want to see how these energy policies work out in the long run.
Energy is big money. Anything that disrupts the status quo is immediately vilified. It got to the point that our major party at the time lead a “gas fired” green energy policy.
They got voted out, the new guys implemented a subsidised home battery policy. In the space of a year it has destroyed the economics of a carbon fuelled grid. All the economics pointing to renewables being the way forward were proven correct.
Everyone gets their back scratched. If you’re a consumer and you invested, you have a massively discounted personal power station that generates you money. If you operate the grid, your transmission costs and peak transmission costs went down. General consumer prices went down because of a reduction in peak electricity use driving up retail costs. Everyone in won.
Except big gas, who cannot turn a profit on a renewable success story. Don’t let rhetoric become your reality.
It's absurd how the propaganda against green energy on certain political spectrums essentially turned into, "It's all money/cash grab for green energy companies!"
As if the established industry has no incentive in the opposite direction? Everyone is going to act in their best interests in this system, that's how it's designed, the whole point of government is to mediate the system in the best interests of the whole.
RE "...how the propaganda against green energy...." One problem is statement are made , that depend on certain definitions , where the definition is not what a reasonable person would expect. Such as statements about costs of renewables. How exactly is the cost of renewables measured.?
I'd have to say there are lots people who do not believe the government is acting in their best interests. Its one of the reasons why Pauline Hanson's "One Nation" party has gained a lot of traction.
> Except big gas, who cannot turn a profit on a renewable success story.
If only.
Chevron’s Gorgon project will be ongoing for decades and making money hand over fist flogging LNG out of Australia to overseas customers.
It'll be a massive source of previously sequestered CO2 and claimed to be "the largest CO2 sequestration project on the planet* .. with the minor catch of (re-) stashing away a very very tiny portion of the CO2 released. Something, something, not let it out in the first place?
Still, could be worse, at least they're not planning on fracking the Kimberley.
Oh, wait: https://www.youtube.com/watch?v=Rh9IkUUgaww
The name is actually "Gorgon"? So a gas company named a project after a legendary monster.
I'm not sure how to feel about this. CO2 sequestration sounds good in theory. But...
RE " Everyone Wins "
Well the tax payer, would be paying the 30% subsidy for Home Battery deployments, which I understand is severeral $Billion 8.5 AU at present.
One group of loosers will be people who can not afford to participate in this program. It would seem another group would be users who cannot shift their electrity demands from peak times when electricty will become quite expensive.
The comment is factually correct, It's a inconvenient truth. Battery deployment is not free.Someone (the taxpayer) pays.
I suspect after they are quite widely deployed. Government could change conditions in future that they must be available as virtual power plant (VPP)
There has been a rush on these since people want access to cheaper power and they believe to avoid the power increases to come. That there will be big increases in power , especially during peak time is the forced introduction of smart meters which allow easy tariff change etc ....
RE "....Look to Australia if you want to see how these energy policies work out in the long run......"
The timeline is several years
> General consumer prices went down
What now?
Where in Australia have consumer prices gone down?
> because of a reduction in peak electricity use driving up retail costs
I don’t understand what you’re trying to day here.
https://www.abc.net.au/news/2026-05-26/power-prices-fall-in-...
> Power prices to fall for most customers, with bigger drops for businesses
> What is driving [lower prices] is a reduction in the cost of producing electricity. We've seen a lot more batteries and solar systems come into the electricity market in the last 12 months. They've been making the market much less volatile," Ms Savage said.
> "We've not needed as much gas and hydro generation in the evening peaks, and that's what's really cut that cost of wholesale generation."
That is an article about a prediction, not actual data about a drop in price.
An article from a month and a half in the past.
Still waiting.
RE "....consumer prices went down...."
A bit misleading comment .. Yes , but after prices went up from renewables
Yep, exactly. Due to the deranged Net Zero policies of successive Australian federal and state governments, Australia have experienced skyrocketing electricity prices, so any small decrease in residential and industrial prices now is effectively a shit sandwich.
The overall American economy is going to pay the price in the long term. And current foreign policy is reducing trust in America as a reliable and stable partner, which will make things even worse.
Well yeah, it's basically locked in now that in 50 years America won't be a world power any more. It'll be in a similar place to Canada at best.
>in 50 years America won't be a world power any more. It'll be in a similar place to Canada at best.
What's going to replace it? Rapidly aging China?
https://www.economist.com/cdn-cgi/image/width=600,quality=80...
https://www.economist.com/cdn-cgi/image/width=600,quality=80...
https://www.economist.com/cdn-cgi/image/width=600,quality=10...
> What's going to replace it? Rapidly aging China?
Using the age profile of a country with over a billion people to dismiss its geopolitical power has no credibility.
More appropriate to point out their (our) lack of domestic chip production and outsourcing of critical technologies.
>Using the age profile of a country with over a billion people to dismiss its geopolitical power has no credibility.
The first two charts are for GDP, which correlates pretty well with geopolitical power. Moreover if we're going by population, wouldn't that imply India, and then later Africa are going to be superpowers?
GDP is subtle broken in many ways
A "in general" correlates with geopolitical power doesn't necessary mean you can point at a specific case look at their GDP and come to a good conclusion. Especially not if it's as much of a special case as China.
> wouldn't that imply India, and then later Africa are going to be superpowers?
which might actually happen. If the US falls further from grace and the situations stays like that for a few decades, or the US and another huge power (e.g. China) fall into some costly conflict then India or a union of African countries have a serious chance to become a world super power.
---
=> But more relevantly this discussion is going into red hering argument direction:
- No one argued that you should go _only_ by population, the argument is, that if a country with a gigantic population shrinks hugely they still have enough people left to wage large scale wars.
- Similar no one (should) argue that GDP is by itself meaningful for such an assessment, it's one imperfect not supper reliable statistical point. Especially given how things move wrt. dark factories (i.e. fully automated => no light one as no human there), AI and similar. Yes, if GDP is supper low or high that says a lot, but a lot is not everything and a average GDP for decade long trends in a word looking to face a major time of change means not much in any direction.
- Similar we are speaking about decade long developments, current military strength ist also just another non fully reliable indicator. E.g. Germany post WW1 had close to actual 0/no military strength, and then there was WW2 just a few years later...
So going by one statistic (GDP) is bad, but going by another (population) gets a pass because... China's "special"? I'm sorry, but that just sounds like a special pleading. I don't like this because it just turns arguments into a fact-free zone where objective facts don't matter and the only thing that does is handwavy vibes from either side. Is China going to be the dominant superpower? The supporters can handwave endlessly about robot armies and the superiority of an authoritarian political system. The detractors can handwave endlessly about the demographic crisis and the fragility of an authoritarian political system.
I litterally wrote a comment saying going by a single statistic is not sufficient.
and it's not China being hadwaving "special" but the measurement you are applying really not working well for a country which is only half capitalist and half planed economy. Which is a separate argument from dark factories which is about (non low end) production in China getting increasingly decoupled from human worker and in turn is way less affected from a population collapse then it would be in other countries. This has nothing to do with robot armies, an argument you just made up, even if China would lose half their population they still had more people and in turn can theoretically man a larger army _even if they have a by % much larger population collapse_, they don't need any Robot army. Similar no one said their system is superior. Or do you mean the US, as a two party system with an extremely powerful president is something some would call semi authoritarian. Either way China having done some pretty effective decisions in recent years in some areas doesn't in any way mean it's the better system. Like a Monarch can do grate decisions, but Monarchy is still is a very bad system.
Either way I also can't follow you how you jump from a listing of rough facts to there supposedly being a fact free zone.
What there isn't is a single simpel conclusion.
I mean it's a discussion where a single bad decision of a larger power can easily cause a unpredictable chain reaction which completely changes what are and aren't likely outcomes.
The world is inherently complex and not everything has a simple black/white conclusion, actually most things don't have one. You can only look at all available metrics and do a guess what the future will bring.
> wouldn't that imply India, and then later Africa are going to be superpowers?
Conventionally, a continent isn't a superpower. A country is.
Africa is not a country.
You don’t need a numerical advantage to dominate if you have a technological / organizational advantage. The British empire was outnumbered 10 to 1 and still managed to control a large portion of the world for a while.
US government is working overtime to make sure legacy technologies stay around while the rest of the world is investing into the future.
> The British empire was outnumbered 10 to 1 and still managed to control a large portion of the world for a while
It also had the support of a lot of the population of the areas it controlled and at least toleration by the majority. Most of its colonial civil service and military were people fro the colonies.
>US government is working overtime to make sure legacy technologies stay around while the rest of the world is investing into the future.
"Rest of the world"? Europe isn't even a player in the AI race. China might be better placed, but they also lag the US in terms of model development and capital deployed. They've been catching up, but that's not the same as "[American will] be in a similar place to Canada at best". That would require a lot of things to go right for China and/or a lot of things to go wrong for the US.
AI is purely a virtual world play. It may (or may not) make it easier to write code, reply to emails or Slack messages or create PowerPoint presentations. In the real world what matters are energy, natural resources and making (manufacturing) things.
AI will not make it easier to build a house, to harvest a corn field, to assemble a car. AI will not improve weather forecasting accuracy- this is already pretty much 99.999% solved. So AI will not improve standards of living.
All this to conclude that China’s standards of living will improve while America’s will stagnate (or worse)
But sure, go ahead and enjoy those AI videos-some of them are even funny.
>>US government is working overtime to make sure legacy technologies stay around while the rest of the world is investing into the future.
>"Rest of the world"? Europe isn't even a player in the AI race.
I think the legacy technology he is referring to is the more about the electrical generation method rather than the AI use of that electricity.
Yes, all the technologies of the future demand a huge increase in energy generation that fossil fuels can’t satisfy. In the long run (decades) the US won’t be able to stay in the game with this kind of anti-scientific policies.
> What's going to replace it?
I'm not sure if I'm mis-reading your comment, but:
The idea that you "need a single extremely powerful country to replace it or the US will stay in its position" just doesn't add up. I'm not fully sure if that is what you meant but I have seen that argument a few times.
_There doesn't need to be a replacement._
For most of the worlds history there wasn't a single leading world supper power, just regional super powers, or supper powers in specific sectors.
And if we speak about who could become a supper power _after the US stops being one and there isn't one for some decades_. Then there are a bunch of candidates.
E.g. China has a population collapse issues, but due to it's cut throat ruthless government and total population number and lead in factory automation might be from all countries with similar problems by margin the most likely to overcome it without power collapse.
E.g. the EU might have many issues (as it's members has many issues) but the general concept has proven hugely successful. And what people forget is EU is the cherry on top of a huge stack of treaties which can be extended to 3rd parties independent of full EU membership. Like e.g. Canada, which doesn't plan to become an EU member, but might join some of this treaties as result of the non stop hostility from the US. Same for military treaties between NATO members which are independent of NATO like e.g. JEF, which would likely be open for letting Canada join. Anyway way to many words. But the end result may be a world supper power which is not one entity but a complex lossy coupled network of countries.
Then there have been some other cross country unions and countries, which want to gain super power relevance but have so far been too irrelevant due to a combination of corruption, bad luck and sabotage by existing powers etc. But if we speak about going for a few decades without any leading super power and maybe prime canidates harming each other (e.g. a US <-> Chineese war) then they do actually have a chance.
>The idea that you "need a single extremely powerful country to replace it or the US will stay in its position" just doesn't add up. I'm not fully sure if that is what you meant but I have seen that argument a few times.
The OP goes further than merely claiming that US will lose its spot as the top power. It claims it won't be a world power at all:
>in 50 years America won't be a world power any more. It'll be in a similar place to Canada at best
For reference, Canada ranks #11 in GDP, below Russia and even Brazil. On military spending it ranks #16, below Poland and Spain. It doesn't have nukes. To suggest US will ever come close to that in 50 years is laughable.
> claims it won't be a world power at all:
which is very much possible
50 years once just a blink in human history is today enough to have 2 major industrial revolutions
nukes are not an argument for anything but main land defense from invasions
Look Russia, it was considered on of the most powerful countries military wise and cleanly in the top 10 economically. Many treated it as a 3rd place behind USA/China. It probably will be neither military nor economically overly relevant in 10 years. Through comparing a war economy with peace economies by most normal usable standards (e.g. GDP) isn't very meaningful. Yes it has nukes, no one will invade it (in a traditional large scale sense out side of some border conflict). But it's economy looks like the kind of war economy which leads to a complete economical collapse once the war ends (i.e. burning core substance) end while it's military could be very experienced now, by treating soldiers as highly disposable its unclear how much that actually is true.
Now look at the US, it has lost the majority of soft power thanks to incompetence (and it was maybe the nation with the most effective soft power). It has alienated most of it's allies, especially it's closest allies. Both from a military and economical POV. It's highly dependent on China, and tries to change that have shown very mid to bad outcomes. On the other hand China is increasingly less dependent on it (and the EU) and increasingly less attackable by sea blockades and other non-hot war counter measures the US had. It's behind on automated production. Countries are increasingly less willing to tolerate US bases on their ground which are essential for it being able to project power. The economy is increasingly dependent on a few pillars, which are strong but can crack anyway. The might claim AI leadership, but practically it looks like the price they payed to get there is one which not only will not pay of but can not pay off, it's just way to high. War is also increasingly going back to ways where even if you have superiority the cost of waging it is still severely high, in such a situation you can't force leadership by military force, if you try it likely will end similar to Germany in WW1. Lastly the internal instability Trump has brought by dismantling checks and balances and arbitrary gutting of programs which turned out pretty essential for stability is bad enough to argue things like a putsch series or civil war being _realistic possibilities_ (not guaranteed at all, but possible where it wasn't realistically possible before).
If you put everything together the US dropping temporary out of the top 10 of overall most powerful countries is, not very likely but, very much possible.
It does have absurd geographical advantages, and nukes to make sure it keeps them basically for ever (as long as it doesn't splinter which isn't very likely tbh.). So the drop would be most likely be temporary at most.
Still it is possible, it's the price to pay for first squandering all trust/goodwill gained from WW2 over a few decades of greedy, egoistic, egocentric, ethical catastrophic politics. And then going complete "unpredictable, untrustable, instabile, high risk, publicly considering to start war with NATO or at least half of the EU (JEF+some others)" in recent years.
>> nukes are not an argument for anything but main land defense from invasions
But it is not an argument even for that. Imagine Russian tanks are crossing the border towards your capital. Will your government use nukes is such case?
lol… you don’t know anything about the Chinese government if you think they aren’t already preparing for this. And the difference between them and us is that when they prepare for something, they actually prepare for it with real state capacity and planning. You know they are building armies of industrial and humanoid robots for exactly this reason?
Also, sorry but the economist hates China with a passion, you really shouldn’t believe a thing they say on the topic.
>Also, sorry but the economist hates China with a passion, you really shouldn’t believe a thing they say on the topic.
Note I'm just linking charts and figures, not their editorials. It's ironic that you accuse "economist hates China with a passion", but the best refutation you can come up with is some handwaving about future robot armies. How do we know you're not just hating US (or loving China) with a passion?
Isn't the US also approaching a peak?
https://www.reuters.com/graphics/USA-POPULATION/AGING/gdpzao...
There's so many ways to slice the data, I'm not sure there's an apples-to-apples comparison between US and China. However, it's fair to say both economies face demographic headwinds.
> It'll be in a similar place to Canada at best.
If Canada was a US state it would rank around 4th or 5th in terms of economic size and power. There's a lot of doomerism, but I seriously doubt the US is going to lose its top 4 or 5 economies in the next 50 years.
If you want to go really doomer, then I think it's actually more likely that Canada would become a US state instead of the US economy shrinking to the size of Canada.
But this is all nonsense talk anyway.
Speaking as a European, current foreign policy is not "reducing trust in America as a reliable and stable partner". It has already reduced trust to rock bottom.
It's hard to see how much worse it could get, short of outright invasion or allying with Russia.
Just this week, amongst other news:
* America's useless adventures in Iran have used up so many Patriot missiles there aren't any left to defend Kyiv against Russian missiles, while Iran is still firmly in control of the Straits of Hormuz.
* Meddling in Greenland went up a notch with landing equipment to start oil drilling.
* State Department funding MAGA-oriented "free speech" disinformation campaigns in the UK.
* The revival of Trump's tariffs under the hypocritical guise of guarding against slave labour. Especially when the US prison system is such a large domestic source of slave labour.
Zero trust left here. Zero desire to visit the US ever again.
I feel it is hard to grasp for an American how much perception of US shifted in eu. This is just anedoctal of course, but a famous Italian singer famous for having written a song about American and sexual liberation decades ago, wrote an updated version. The former is "America", the latter is "America INC." You can guess which one depicts America in a bad light.
I wouldn't have expected a famous singer to make such a song, given that America was always portrayed in italian pop music with a positive image, and negative views were relegated to mostly political groups or singers with political themes
Why do you think Americans don't grasp it? There are really two groups here:
- MAGA who doesn't care what Europe thinks - not-MAGA who are just as shocked and disappointed as everyone else on the planet
Simply put, I think Americans don't grasp it because even for us Europeans it is sometimes surprising how repentine and fast the shift has been in the regular Joe's brain - I'm not saying that Americans are stupid bla bla thing -, I'm saying that it is normal not be as up to date with some phenomena as people living where that phenomena happens.
Also, at least the Italian government is still dancing the tip tap regarding a position on America, not taking any definitive stance or doing anything significantly different than before, which is a big rift wrt what people want. Only exception for defense expenses
I don't think the first group understands how important Europe is to America's status as superpower, while the second group operates under the delusion that the world will go back to business as usual after the next presidential election.
Only the elderly who are blind to the decline and can’t even picture a world order outside other than Pax Americana believe we will go back to business as usual. The rest of us realize it will take decades of stability to gain a semblance of the trust we used to be granted. And in the mean time the rest of the world will forge ties that don’t include us. We will be weakened on the world stage compared to status quo prior to trump for the rest of my life I’m sure.
You think there's gonna be a next one? He's installed his own people in all the right places, and got his own Praetorian guard.
Plenty of us in the second group don't think the world will go back to business as usual.
I'm sorry, friend.
There is a lot of that with the population but I don't see it at the leadership level yet. The top politicians are all still desperate to go back to the good old days of euro-atlanticism (and all the cushy positions it gets their friends & family), there's also the problem of all of their capital happily sitting and growing in fortune 500 companies.
> It's hard to see how much worse it could get, short of outright invasion or allying with Russia.
Why would the US ally with Russia, who is a Chinese and Iranian ally?
That's just uncritical thinking.
The United States is the single largest bilateral provider of direct military equipment, intelligence, and defense funding to Ukraine (a European country) against Russia.
> Why would the US ally with Russia, who is a Chinese and Iranian ally?
I don’t think the comment above you is suggesting that is a possibility, but rather saying “things are so bad that the only imaginable way to get worse would be <outlandish unimaginable scenario>”. In other words, it seems to be hyperbole.
Perhaps, though my reading between the lines is that they think trust is so low because the US won't help defend Europe and might invade Europe, even though, despite rhetoric, all actions point to the opposite happening.
It is, as the USA has a large population. We can be more nuanced.
By absolute dollars, single country: US > EU institutions > Germany > UK
By absolute dollars, bloc vs. bloc: EU/Europe combined (~$237bn+) now exceeds the US (~$130bn), especially after 2025
By per-capita/GDP burden: tiny front-line states (Denmark, Estonia, Lithuania, Norway) far outstrip both the US and the larger Western European economies — the US actually ranks closer to the bottom of the pack on a relative-effort basis, behind Germany, the UK, and virtually all Nordic/Baltic states
Between February 2022 and January 2025, average expenditure per European resident stood at $394; by April 2026 that figure had risen to $618, driven by the total spending by European NATO countries on Ukraine exceeding $237 billion, including $96 billion contributed just since January 2025.
Average European (across NATO's European members) $618 per person (2022-2026)
United States (my estimate: ~$130bn total ÷ ~340m population)≈ $384 per person
That is an indisputably rough list, and I'm sorry that's your conclusion.
It's tough being here and feeling powerless to stop it. Beyond voting, campaign donations, judicious choice of employment, etc.
Anyhow, you're welcome back should you change your mind. In my book, anyway.
> feeling powerless to stop it
You guys are the meekest sheep to ever been led astray by a howling, undisguised wolf.
What more does this regime have to do to warrant massive, historic, disruptive, non-peaceful protests in the USA?
Have you seen protests in France? Bringing the country to a halt is the least you have to do in order for us to ever have faith in the USA again.
Until then, nothing any US citizen can say online will do anything to restore trust. If you're not rioting, you're complicit. It's as simple as that.
I admit I haven't followed the French protests as closely as I should have. Do they shoot protestors in the face with teargas canisters there too? Maybe they do. Do those protestors face the real possibility of medical bankruptcy if they need extensive care? Does France enforce laws that protect striking workers?
Where do you live? Does everyone have an AR-15 in their closet? You do not seem to grasp the reality of our situation.
The problem with the US is not the regime, it's the people. We elected Trump a second time with ~50% of the vote after seeing four years of his complete buffoonery. His approval rating is still around 40%. If he were somehow to run again in 2028 (never mind it being unconstitutional), there's a pretty solid chance he'd win. His party still backs him 100%, and nobody has any idea whether they'll maintain control of Congress after the midterms, because they still have enough support to plausibly keep their majority.
It's not about being too meek to fix things. It's about half the country wanting what's happening.
If you don’t want to fight for the things you care about, you don’t deserve them. Giving up before even trying anything is completely absurd to me.
The environment's already lost. We've passed the 1.5ºC deviation mark from the pre-industrial global average, and we're on track to hit 3ºC in a few more decades.
The excess deaths and human misery (both that which is increasingly clearly ongoing, and we have to look forward to) are going to be tragic.
Yes, but there's still a range of possible outcomes, depending on what we do next.
You'll forgive me for harbouring some pessimism about current world leaders' climate strategy.
I didn't say we'd choose wisely. But since there is still a range of possibilities, I'm not giving up on the whole thing.
This week, the US government paid a wind farm company $1.2 billion to stop building wind farms. It also made solar inverters illegal last week.
> It also made solar inverters illegal last week.
That’s overstating it. A couple clarifications.
It banned foreign (edit) made equipment because it’s known for cybersecurity issues. It was the FCC that restricted new foreign-made connected inverters over grid cybersecurity concerns. [1]
1. https://www.fcc.gov/document/fcc-adds-foreign-produced-power...
> This week, the US government paid a wind farm company $1.2 billion to stop building wind farms.
The $1.2B is real, but it wasn’t payment to stop construction. RWE agreed to surrender three offshore wind leases and settle related claims. Anyways, RWE’s three leases were in very early development and weren’t expected online until the 2030s. [2]
You can still think both policies are bad. But those are different claims.
2. https://www.reuters.com/legal/litigation/rwe-says-reached-12...
It banned all foreign made solar inverters i.e. all solar inverters.
That’s true. China dominates globally but I found that the US installed European and Japanese models. There is a big supply chain concern and set of reviews happening all across the entire government sector. Energy is a national security concern. [1]
1. https://www.energy.gov/cmei/systems/solar-photovoltaics-supp...
It's bad policy.
If one is trying to attack the threat you claim, one might ban all internet capable inverters.
To ban all, even fully offline ones, means the threat you describe is an excuse.
Your other comment got downvoted to death (not by me) so I can't reply to it, so I'm replying here. Sorry the downvoting happened; your comment wasn't incorrect. You're right that inverters with zero connectivity whatsoever are not banned; my comment was overly broad. That said...
The ban covers any inverter with any wireless communication whatsoever. Bluetooth is specifically called out, despite necessarily being local-only and not enabling an internet connection.
I suppose you could make the argument that with Bluetooth, the app that communicates with it has internet access and thus it's the same thing. However:
The ban does not cover a power inverter with no integrated comms, but has a serial port into which is plugged a "sold separately" wifi dongle. Neither does the ban cover an inverter with an ethernet port. It covers solely wireless communication.
"connected power inverters produced in foreign countries"
I think home devices that phone somebody-elses-home are a cancer anyway.
[1] https://docs.fcc.gov/public/attachments/DOC-423682A1.pdf
Even if I assumed maximum malice, how are these inverters going to perform a coordinated attack on the grid?
> Even if I assumed maximum malice, how are these inverters going to perform a coordinated attack on the grid?
They’re network-connected and remotely controllable. The concern is compromise of the vendor/cloud infrastructure allowing large numbers of inverters to be manipulated simultaneously.
Or even Trojan horse style malware is a possibility I suppose (Stuxnet).
That’s basically the threat model that they are using. If people want to argue with that, then that’s something to poke at
AC power needs it's sine wave aligned or it's very bad. If a huge chunk of inverters all start dumping power onto the grid that isn't synced, things break.
You can't just "not sync the grid". I think what you mean is if all the country's inverters that are curtailed stop being curtailed and pull the grid out of frequency because they control so much power. Note that every inverter as well as every traditional power plant and many substations will disconnect themselves long before reaching 59 or 61Hz - the most you achieve like that is a blackout.
Do people really believe this stuff? Wow
Modern data centers can be designed with near-zero direct water consumption. It isn’t fundamentally necessary to consume huge amounts of water to cool computers. What’s the story with those new data centers? Do they actually choose to do evaporative cooling that people are complaining about or people are complaining about problem that doesn’t exist?
They choose evaporative cooling simply because it's cheaper than a closed loop system for energy use and land area.
How? In most locations for much of the year you need to reject heat either by evaporating water or using electricity (more water) to run a refrigeration cycle.
It doesn’t matter what the actual reality is. People think that data centers waste water, and they don’t like that.
The beef from a single hamburger uses far more water than a year of ChatGPT usage. Few try to reduce consumption in favor of more efficient foods by comparison. It’s about perception.
Exactly, this is all a symptom of underinvestment in grid capacity, and datacenter companies having to cobble together whatever electricity source they could get their hands on at the last minute.
It's a complete failure of the state in having any kind of foresight.
China, in comparison, has around 2X excess electric production capacity in their grid. That's what made mining Bitcoin, and now ultra-cheap LLM inference, economically viable.
A country is but the sum of its people - and this american people have gotten what they voted for.
Nice opportunity for the rest of the world, though.
> ... and this American people have gotten what they voted for.
Probably one of the biggest flaws of a Democratic system is that a voting majority can mandate a situation that harms literally everyone against the will of everyone else even when those harms are well documented and absolutely provable.
"The majority voted for it" isn't a great answer for the question "Why exactly are we letting the bus driver drive us all over this cliff edge to our fiery deaths?"
"Stop the bus! I want off!"
"Nope. The vote's done. You're stuck with the result."
"Well, what if I voted against driving over the cliff edge?"
"Nope. Majority rules. We all die now."
Bear in mind the US electoral system is far from the best that can done.
Not all democracies are equal and the US system is barely "democratic" - it's spiralled in a two party not broadly representative Hotelling's Law swampland.
> Those in power seem to think that a compute advantage is necessary to maintain economic advantage. Despite this, the voters universally hate data centers, so much so that both NY and TX have data center construction moratoriums.
I don't understand what "despite this" is doing here. People have thoughts about what good/bad policy should be, independent of what good/bad policy should do all the time.
This is not an argument for or against data centers, but rather against populism. People have too many feeling about too many things, that should simply should not be decided by feel (and, sometimes, not enough about things that should).
The sense of urgency is strange and feels extremely forced. There’s no way this is all so important that all other concerns automatically go out the window so projects can complete construction yesterday.
It really wouldn’t hurt if the developers involved tried even a little bit to be good neighbors and to select sites where they’re not going to be a nuisance and net negative to adjacent communities at minimum.
The blatantly extractive, externalizing mindset of this all reminds me of the attitude that mining companies have taken towards Appalachian states for over a century at this point.
Given local grassroots opposition, the data centers are starting to move into less regulated countries where I would gather building them raises the local standard of living, but presents the new challenge of finding people qualified to run them.
The entire planet will lose as ecosystems die out, food and water supplies run low, refugee crises and global conflicts ramp up.
But hey, line go up!
Don't worry; republicans are way ahead of that refugee crises by generation fervent far right ethnic nationalism globally, independently of course.
This is just pure histrionics, you have no understanding whatsoever of the scale of data center water consumption, electricity consumption, new capacity coming online, etc.
Climate change was already a crisis before the current wave of data center construction, but don't let that stop you from leaving insulting comments.
How much water does a DC use?
Not much. 0.04% of US water production. For context, that's about 3% of the water consumption of our golf courses, and less than 1% of what we use to water our lawns.
Get off my lawn.
All US data center water usage is less than the total water used for golf courses or the total water used to grow alfalfa. Less water used in evaporative cooling means more electricity is used for cooling given that the efficiency almost doubles if you add evaporative cooling to a chilled water loop.
It’s bad to pump non-renewable fossil aquifer water in Arizona for evaporative cooling, but using cooling towers in an area that gets 30” or more of precipitation per year is probably not going to hurt anything.
The question is does DC building bring added costs to local communities. My small town has a backlog of home builders waiting for permits because of constrained municipal water capacity. We are still paying off the bonds for the last capacity we added. Golf courses/etc are already factored into current costs. Additional new use brings significant cost (bonds to build new capacity), funded by increasing expenses on the entire community, not funded by those who financially benefit from the new build outs.
The issue isn't comparative usage to existing hookups, it's new cost burden on the local community where a data center is built. Which is why DC propogandists switched to the golf course measurement, it confuses the original concern, the cost to the community of new DC hookups to municipal water or the very limited amount (of expensive to the community) available hookups that were intended/funded by the community for other purposes.
Given the political will, that problem is straightforward to address with the permitting.
> Despite this, those in power refuse any environmentally aware energy policy nor water policy.
Can you tell me how much water a data center uses? If you have strong opinion on the construction of data centers I hope you can answer this basic question about data centers. (Many opposed to them cannot)
Depends on the cooling system, but around 1 to 9L per kwh. [1]
A lot, but not all, of these datacenters are using evaporative cooling, which is what kicks their water usage so high.
So a 40MW datacenter will consume around 4 million liters of water per day if it is evaporatively cooled.
[1] https://www.fwpcoa.org/content.aspx?page_id=5&club_id=859275...
As a comparison, one acre of alfalfa growing year round in California uses about 5*325000=1,625,000 gallons of water a year for four harvests, 4400 gallons of water per day per acre.
There’s around a million acres of alfalfa grown in California for a total water usage of 1,625,000*1,000,000=1.625×10¹² gallons (or 5 million acre-feet) per year.
365 million gallons of water per year for one data center is 1000 acre feet of water. California’s alfalfa crop uses as much water as 5,000 of these data centers.
That the only nation which sent man on the moon uses acre-feet as a unit of volume blows my mind
Oh sure, I can help.
1000 acre feet is approximately 40 hectare meters. Or, you know, about 1.2 billion liters.
Over on this side of the ocean, we’ve got freshwater lakes larger than entire European countries. There’s a lake over here, it has an island on it. That island has a lake, which has an island in it. And there’s a lake on that island.
We need units of measurement that represent large things. If you don’t, that’s ok.
1.2 billion litres = 1200 megalitres = 1.2 gigalitres
Works well in places that need to represent things large and small. Europe, Canada, Australia, and pretty much everywhere else.
But enjoy your cubic furlongs and tablespoons.
Do Alfalfa growers traditionally use limited (and funded at large expense by the local community) municipal water hookups to grow their crops? Are alfalfa growers asking for new water hookup permits to access these resources?
Because DCs are, and that is the concern local communities have. Not random comparative statistics whose purpose seems to be to take valid local community concerns and pretend they're just a joke. How much do alfa growers in California add to my water bill? How much will new municipal water hookups by new data centers impact my water bill? One is a valid concern. One is just a BS attempt to discredit the valid concern. Because that is how this shady industry operates. On BS. On manipulation. And they wonder why they are more and more disliked.
It is a sort of "damned if you do, damned if you don't" situation.
The reason these DC use so much water is because it uses less electricity to evaporatively cool than it takes to cool with a heat pump.
You are trading water scarcity for energy scarcity.
In my personal opinion, the right tradeoff is more electricity. Also IMO, these DCs should be required to build out solar + battery + wind offset plants which supply their energy either directly or indirectly by selling to the grid operator at a discounted rate. If you want 40MW of power, you need to deploy 40MW of solar (not peak, but yearly average output) and a 400MWh battery plus all the grid hookups required.
The end result for residence should be lower energy bills, not higher bills.
> Do Alfalfa growers traditionally use limited (and funded at large expense by the local community) municipal water hookups to grow their crops? Are alfalfa growers asking for new water hookup permits to access these resources?
No, you already know they use flood irrigation.
Why is a municipal water system allowing a data center connection that will strain their distribution network? They’re well within their rights to say no, Mark Zuckerberg isn’t holding a gun to their heads.
> Why is a municipal water system allowing a data center connection that will strain their distribution network? They’re well within their rights to say no, Mark Zuckerberg isn’t holding a gun to their heads.
Deep pockets have a way of bypassing practical concerns. [1]
Zuckerberg likely isn't holding a gun to their heads, instead it's more likely he's putting wads of cash in their pockets.
[1] https://www.texastribune.org/2026/03/08/texas-corpus-christi...
Assuming that it's surprisingly little and without pollution concerns, then an acceptable and enforceable policy should be a very simple request, no?
Yes I agree.
So even when xAI (which is now Tesla? Or SpaceX maybe?) was "operating dozens of unregulated gas turbines that worsen the air quality in one of the most polluted parts of the country"[0] this didn't place them above this new Amazon not-even-built-yet data center? Are they competing who can destroy the earth the fastest or what's going on?
- [0] https://techcrunch.com/2026/05/20/musks-xai-is-being-sued-ov...
It seems wrong that Amazon should get in trouble for this when it was XAI that began this new race to the bottom.
Peoplecheered xAI on and admired how quickly they were able to move, without much regard to the methods or legality. IMO competitors had no choice but to respond in kind when they see XAI not just getting away with it, but also getting a pass on legal issues.
You think Amazon shouldn’t get in trouble for this because someone else did it first?
> It seems wrong that Amazon should get in trouble for this when it was XAI that began this new race to the bottom.
Who gives a shit who started it? The ones who continue engaging in it, should clearly be punished as they're working against humanity's best interest of keeping this planet survivable.
If you're one of the largest companies in the world, constantly brag about how environmental friendly you are and how much you care about "sustainability" or what the trendy word is today, and you end up building a data center that sets a record in pollution (either directly or indirectly), you're doing something massively fucked up and are misaligned with the rest of the people on this planet.
> Peoplecheered xAI on and admired how quickly they were able to move, without much regard to the methods or legality. IMO competitors had no choice but to respond in kind when they see XAI not just getting away with it, but also getting a pass on legal issues.
I guess this is what late-stage capitalism gives you ultimately, companies who feel like it's fine doing whatever because competitors are also doing whatever. No one I know have been cheering for anything like that, I suggest if you have people cheering about stuff like that near you, get far away.
Pretty sure it’s “is”, they were operating gas turbines for power and continue to do so
So where is the police? They're running turbines they haven't been granted permits to run, if Average Joe did so the police would show up and literally grab the hardware from him, one way or another, or disable it at least. What the fuck are they waiting for?
Where are you getting the information that it's unauthorized (or needs to be authorized)?
> Where are you getting the information that it's unauthorized (or needs to be authorized)?
https://www.selc.org/press-release/naacp-asks-court-for-emer... and the previous link I referenced.
One would think if there is even a suspicion (even if it hasn't been ruled in court yet) about large-scale harmful toxic emissions, the police and relevant authorities would at least quickly do some samples or what not, to figure out if they need to put an immediate stop to it until the courts ruled either way.
But seemingly it's still ongoing and no harm no foul?
Do you know the difference between civil law and criminal law? The police don’t care if they’re not violating a criminal statute. If you want to test how it works, build a brand new house in a city without pulling any permits.
I do indeed. Do you know that severe violations of civil law sometimes amount of up to criminal law, especially if there is evidence of criminal intent or "knowing" violations of the law but proceeding anyway?
This is the case for lots of things. Negligence can turn into "reckless disregard for human life", breach of contract could amount to fraud, and many other cases.
The NAACP claims don't amount to facts.
Indeed, but they do amount to "suspicion" and are worth investigating instead of saying "Lets wait years until the courts decided if it is or isn't actually polluting" and not shutting it down until they've clearly demonstrated it being safe or not.
All this AI doomerism is hiding the obvious and immediate harm that will cause a mass extinction event. Funnily enough people at Anthropic are predicting the end of the Anthropocene, completely absolving themselves from any responsibility for the actual reality that the Anthropocene is defined by human made geologically noticeable changes. Like causing a global climate catastrophe.
It's possible to have more than one existential risk at a time. We even have a third, since nuclear war doesn't seem completely impossible these days. We probably shouldn't ignore any of them.
Nuclear war is hardly an existential risk.
An all-out global nuclear war (all nuclear nations landing 10% of their warheads in populated areas) would be a disruption comparable to COVID at best.
Can you elaborate on this? I’m not an expert at all but this seems so far from true I’m struggling to understand if you’re serious.
Wouldn’t we expect tens of millions of casualties, nuclear fallout rendering large swathes of countries uninhabitable, and ensuing total societal collapse once the globalized western world is separated from their foreign means of production?
Of course you're correct. Here's a 2019 simulation of escalating full-scale war between the US and Russia, done at Princeton based on what's publicly known about actual force postures. Immediate effects would be 34 million dead, 91 million total casualties, and 30 major economic centers destroyed on each side.
https://www.icanw.org/new_study_on_us_russia_nuclear_war
Fallout would cause significantly more casualties. Then there are the environmental effects. Science on that has evolved over the past few decades, so wikipedia is a good starting point for an overview.
https://en.wikipedia.org/wiki/Nuclear_winter
Since none of this is certain and we're talking about risks, considering the worst cases seems appropriate, and those involve agricultural collapse, several billion dead, mass extinction, and an ozone hole covering most of the planet. It might not kill off the human species, but modern civilization would struggle to recover.
Mass extinction event already happened though. Most known species became extinct because of humans.
It is the paperclip AI scenario: The AI is cheating humans to produce only one thing - but in reality today, its not paperclibs, but datacenters as "the most important product": So the scenario still holds :-D
The existential risk people are worried about a much shorter timeline - like the next 5-10 years. Climate change is serious but very unlikely to lead to anything like ending the Anthropocene in the next 5 years.
400,000 people in Europe are displaced. Pretending that this is a future issue is tremendously disingenuous. Things don't need to be civilisation-ending to cause a degree of harm that is worth focusing on.
The OP is literally talking about “mass extinction”, it’s not a real concern that climate change will cause mass extinction in the next 5 years.
It should absolutely be focused on anyway. I don’t know how you got the other impression. The US gov paying to stop windmills and solar is insane.
Note that this is about absolute pollution, not relative to power output. Pollution is high, because power output is high. Relatively, for what it's worth, it's just a fairly normal gas plant.
So the most interesting thing to take away from this: There is a lot of demand for more power and in this instance it's satisfied by gas. Which is still much better than coal. Which is still much worse than renewables.
Luckily the Trump administration has just paid over a billion dollars for the second time to stop a renewable energy project...
Mostly because coal is more expensive and slower to build
On this map I can't see _any_ nuclear under construction in the US yet the rest of the world seems to be building many. When a man has a hammer everything is a gas turbine. GEV stock has treated me well but I'd much rather nuclear was winning.
It's a catch22 problem because nuclear is much slower to build than a data center so there is no demand from the data centers. And the nuclear plant doesn't know the demand will be there in 5 years once with gas soaking up demand. So the build it and they will come strategy is risky. Building infrastructure is like a muscle that gets stronger with repetitions. The cost will only come down once some are built but nobody wants to build at the current high prices. Govt help is needed to kick start it.
Forgot to post the link -> https://world-nuclear.org/information-library/facts-and-figu...
DOE is investing heavily in SMRs right now. There's 4 which have reached criticality so far. Chris Wright the DOE head is a o&g political appointee hack, but the career employees at DOE are working on pushing clean energy forward.
https://nric.inl.gov/news/
The world isn’t building nuclear power. It’s simply too slow to build and expensive to be relevant.
That's not true though. China has ~20 plants under construction.
In 2025 China built 696 TWh of new wind + solar power and 13 TWh of new nuclear power. So to a first approximation it's fair to say that even China isn't really building out nuclear power anymore.
Great comment. China is still building Coal and Gas power too, I wonder how they compare.
Most polluting power plant so far
This race to the bottom will have plenty more to give us in the coming years.
Aren't coal power plants waaaay more polluting than natural gas? Not just in terms of C02, but also really nasty stuff like particulates, sulphur, and metals?
It's very trendy right now to get mad at datacenters, but I'd much rather energy come from natural gas than from an existing coal power plant. (Obviously, renewables would be better than either.)
The nasty stuff, yes. Regarding greenhouse gases, natural gas has turned out to be about as bad as coal once you account for methane leaks, and the industry has strenuously resisted any decent mitigation for that.
Different. Burn enough natural gas in enough gas turbines, and significant quantities of formaldehyde and NOx will be created.
Who could have predicted that Microsoft building their own nuclear reactor would be the lesser evil.
Until they apply what that did to Github's uptime
Nuclear reactors are a notoriously clean and safe method of electricity generation.
Ok. A serious question. I frequently hear that wind and solar are the cheapest forms of electricity. Then why are they using gas turbines? Is it because they can generate more power with a smaller real estate footprint?
Texas specifically has heavily politicized and demonized wind power, in spite of having massive successes with it prior to 2016 and being a leading producer of it. Solar isn't as bad, but is still seen as "leftist." The political landscape in TX is vastly more friendly to natural gas.
In other words, they are using gas turbines because the politicians are stupid.
They are using gas because it's more reliable during peak usage. Wind generation in Texas isn't optimal throughout the entire state and conditions aren't optimal all the time. They still use wind, but from a pure numbers perspective, gas infra already exists, the gas is there, and gas is more reliable, so wind isn't always the cheapest form of energy. There are also diminishing returns on just adding more wind turbines.
Gas is also incredibly cheap in Texas. Way, way cheaper than you would think. In some parts, it's negative price - you will be paid to use it. We have entirely too much gas, which is actually a problem because ideally we want to discourage its use.
So yes, wind and solar are cheap. Depending on where in Texas you are, gas is cheaper.
Texas leads the nation in solar installed and solar planned.
How do you reconcile that with your theory?
*utility-scale solar, not solar overall
Utility scale, solar installs, and will be total solar by end of year.
Texas accounts for 40% of new solar capacity installed per year at this point.
It does lag behind an inefficient and expensive rooftop solar, and has no State mandate for homeowners to build it.
That said, I think this is enough to show that Texas doesn't hate solar, so there's probably a different reason for selecting gas to support the data center. If I were to Hazard I guess, it would be stability and demand curve
I think you might want to download new anti-Texas talking points. Texas is number two in the nation in solar power: https://seia.org/media/16020/texas.pdf/
Texas is not at all against renewables. Maybe, just maybe, whatever axes the people building data centers are optimizing on rule out solar/wind over factors other than "renewables are for leftists, so fuck that we're gonna burn gas even if it's more expensive."
> I think you might want to download new anti-Texas talking points.
No need—the current ones are uploaded directly into my mailbox regularly by Randy Weber, my district rep.
15th for non-residential? Sounds like people overwhelmingly put it on their roofs, and politicians avoid it.
Did you miss the 1st utility part? It can be quite hard for nonresidential users to retrofit for solar. Not impossible but sometimes not an easy solve.
Sorry, I do have a bias.
> 11.40% of TX electricity is from solar
This is... Not a huge statement to me. My home state is at 16% solar, and we're calling it inadequate. [0] Its not a 1-to-1, as we have probably 10% of the people, and 1/3 of the landmass. More spread out, different demand, so a completely different landscape.
But 11% is not some victory dance.
[0] https://www.energy.vic.gov.au/about-energy/news/news-stories...
Avoid it to the tune of being in the top third (for now, they're projected to keep growing), behind only blue states that are known for aggressive rollout and/or climates that are overall more favorable (i.e., Texas gets its fair share of sun but not quite like CA and NV, for example)? Is that your idea of avoiding it because politicians want to own the libs?
I often wonder why conservatives have such a problem with renewable energy.
I can certainly believe that oil and gas pay better bribes, and that this incentivises a certain kind of politician who seem to be over-represented among conservatives in the US. But that doesn't explain the widespread dislike among the conservative end of the general population. Is it just groupthink or Overton Window -nudging using social media? Or is there something more psychological going on? Some kind of disgust reaction?
One would think that self-proclaimed "conservatives" would like to conserve the natural world... Long gone are the days of Teddy Roosevelt that's for sure.
Under current administration, I'm not even sure its possible to build out green energy for this sort of thing at scale - this article[1] was on the front page just yesterday
[1] https://news.ycombinator.com/item?id=49208314
The court cases are starting to have effect. https://arstechnica.com/science/2026/08/trump-is-losing-his-...
It's because the permits are easier to get. It's very hard to get people to organize to make renewable energy permitting easier.
In some cases, aren’t the permits not even being sought?
Eg https://www.reuters.com/world/americas/pollution-musks-unper...
Sure, you don't pay millions of dollars in engineering for a permit you obviously aren't going to be able to get. That's a big part of the problem, they're unpredictable.
Because the US has some policies that make those currently more expensive than they should be while. A combination of tariffs, removal of incentives for local producers, etc. has caused prices to be higher than they could be. All of this is because of fossil fuel lobby funded campaigns for exactly this to happen.
The flip side of course is that gas turbines are in short supply and manufacturers are in no hurry to build out production capacity for what is effectively obsolete technology with a very uncertain future. A lot of the projects being announced may never get delivered for the simple reason that there are long waiting lists for these things.
And there is the notion that related investments might end up under water when the few plants that do get built start struggling economically in ten years or so. These are not very attractive long term investments. The short term interest in securing any capacity at all is overriding this. Of the well over a hundred GW of data centers planned until 2035, probably only less than a 30-40% will actually get built; if not less. Michael Liebreich is predicting only about 35GW will be realized. His estimate is that there simply is no more production capacity that can be committed than that.
It will be interesting to see if this Amazon plant actually gets built. For now Amazon can afford to gamble on this and just write it off if it doesn't work out.
If policy changes in a few years, I expect a lot of these projects to get unceremoniously cancelled when it turns out that they are a combination of under water financially and behind on schedule because of component shortages anyway. Also, it only takes a few failed projects for everyone to get cold feet. Premature closures of new gas plants could happen if renewables get cheap enough. Those are already happening in some places.
For a DC it's probably reliability. You need wind to be spread over a large area to even out random changes in power availability. And, obviously, solar power works during the day.
But this is essentially a solved problem even at utility scale, with renewables + battery banks, right? This isn't cutting edge stuff.
Renewables require storage because they are bursty - sun goes down, wind stops blowing. This is more expensive than just burning as much nat gas as you need.
Presumably, the data center usage could also be bursty, right?
More people will be using AI during the (work)day, than at midnight. Then they could do half solar, half gas.
Also "more expensive" when it comes to solar isn't meaningful until you also consider timeframe. Solar panels pay themselves off in a few years, and after that they basically generate "free" electricity. Gas never becomes cheaper or free, every year it will cost money.
I doubt it will be bursty.
First: global usage.
Next: I even wonder why frontiers labs don't sell batch processing and spot prices to keep 100% utilisation. If they do that and I suspect they will it will be 24/7 maxxed.
Imagine this: Fable at 98% discount because but you need to wait. Then imagine: bots that scan your tickets for easy low priority tasks to do when those cheap tokens come.
Let alone research batch compute needs. Training and so on.
One concept will drive this shift: GAAP profitability.
AWS is heavily used worldwide, not just in the US. The same goes for the big AI companies.
Timeframe is accounted for if you look at levelized cost per kWh. The levelized cost of natural gas is 5 to 11 cents/kWh. In the sunniest spots, the levelized cost of solar with enough battery storage to provide 24/7 power is about the same, if you want >90% reliability. The usual four hours of storage is only enough to account for the "duck curve" during the day.
https://www.pv-tech.org/solar-pv-and-storage-cost-declines-d...
https://ember-energy.org/latest-insights/solar-electricity-e...
Solar and batteries keep getting cheaper so hopefully they will be clear winners soon, at least in sunny areas.
How is it more expensive? Any links? Once built out yeah there is maintainance but no need to drill baby drill anymore.
The cost of energy varies widely based on regional factors, geography, existing energy portfolio makeup, etc. Additionally cost is not just the unit price but the price overtime to support a specific SLA.
Texas has very cheap gas; it’s also good for solar. They are very competitive price wise, so it’s kinda depressing they are using gas. But it’s probably a slight price edge plus better reliability.
gas turbines are not intermittent
> I frequently hear that wind and solar are the cheapest forms of electricity
presumably they put zero value on stable production or count subsidies as income
You need a ton of land for solar and no renewable is as reliable as fossil fuels.
HN, where inconvenient truths get you downvoted
A recent discussion
The Gargantuan Lie That is Collapsing The World’s Climate: https://news.ycombinator.com/item?id=49208686
Zero data on Greenhouse gas CO2-equivalent per kWh. A combined-cycle gas power plant is going to have lower impact per kWh than the 200+ coal plants in the US.
I am sure if these guys worked on it they could have better PR. It is like AI is the modern version of the oil industry.
really time for these companies to have to switch to options like Nuclear SMR from providers like Rolls Royce who already have the megawatt solutions ready in the market, which don't release giant volumes of CO2 which we cannot afford.
7 GW! All California load is 35GW right now.
Divest all your stocks from all this AI data-center race bullshit, invest into renewable projects. Worst case, the orange buffoon will bail you out with taxpayer money to prevent the US grid from being too clean or something. Only money talks in the US.
Finally, a structure worthy of its namesake.
1786171869 | Amazon behind private gas plant for new data centers | https://finance.yahoo.com/energy/articles/amazon-behind-mass... | https://news.ycombinator.com/item?id=49219445
1786121329 | Scoop: Amazon Is Behind One of the Largest Planned Gas Power Plants in the US | https://www.distilled.earth/p/scoop-amazon-is-behind-one-of-...
1786014051 | AWS eyes potential data center development at 4.5GW natural gas plant in Pennsylvania | https://www.datacenterdynamics.com/en/news/aws-eyes-potentia...
1786181141 | 'Better ways to use gas': Amazon finances massive US plant for data centres | https://www.scmp.com/news/world/united-states-canada/article...
1786202472 | Amazon plans massive Texas AI campus and an enormous gas power plant | https://www.chron.com/news/article/amazon-texas-data-center-...
1784098800 | Amazon to install hundreds of gas-fired generators at Bucks County data center site | https://whyy.org/articles/amazon-data-center-bucks-county-pa...
1786140583 | Amazon Is Building America's Biggest Fossil Fuel Plant to Power AI | https://heatmap.news/energy/amazon-gw-ranch
1786173935 | Amazon Backs 7.65-GW Gas Plant in Texas to Power New AI Data Center Campus | https://constructionreviewonline.com/amazon-backs-7-65-gw-ga...
1786131006 | Amazon Is Creating the Biggest Pollution Source in the Entire Country | https://newrepublic.com/post/214111/amazon-data-center-bigge...
1784358000 | Pennsylvania set to approve 283 new generators at Amazon data center despite objections | https://www.thecooldown.com/green-tech/amazon-data-center-bu...
1780556400 | What the rise of natural gas-powered data centers means for emission-reduction pledges | https://oregoncapitalchronicle.com/2026/06/04/data-centers-a...
Now no one talks about global warming and climate change.
A lot of new data centres have been announced in Alberta, Canada. Which is a weird location given that Alberta has a derelict power grid that can barely serve the current load, and runs almost entirely on fossil fuels.
Meta, for instance, plans a $13B data centre in Sturgeon County.
https://www.vozpopuli.com/indux/en/alberta-approved-a-932-me...
They all plan on powering them by building massive natural gas generating stations with each. In a free for all race to destroy the environment and doom humanity (aside from your plutocrats who'll live in their bunkers). Until a week or so ago Meta still had a clean energy pledge, but finally they just abandoned that silly greenwashing lie: https://finance.yahoo.com/energy/articles/meta-quits-clean-e...
We are basically in the end game now, and despite that weird guy running through these comments telling everyone how great everything is going to be in a hundred years, things are looking utterly grim.
Oh, Alberta. That makes sense. Land of pissed off coal miners, isn't it?
Much more the land of oil and gas. Alberta no longer burns coal for electricity.
The dominant political party frames natural gas as "clean", because it isn't a dirty as coal; and Alberta natural gas in particular as "ethical", because not Russia or Saudi.
If we reach AGI at some point, it will wonder why we are so dumb
If we ever reach it, it will tell us "you destroyed any chance of survival in your quest to get to me first, dipshits."
It is insane how we're accelerating the climate catastrophe in an ideology-driven race to be king of the ashes.
Capitalism is a death cult, AI capitalism doubly so.
Communism dried up the third largest lake in the world just for irrigation (while starving tens of millions in the meantime).
The irrigation plan was in response to the famine, and it worked, it was the last famine in the soviet union.
also the soviet union was totalitarian, not communist
We only have two options.
No one mentioned communism?
The person you’re replying to did, when he pointed out that a communist regime wasted a ton of water.
I think I get your point; that the inverse of corrupt (I call it fake) capitalism is not necessarily communism, or something reactive. It might just be a system that works better, even if we have not named it yet...
Downvote me all you want. China is doing something right. Nuclear, renewables, whatever the mix, they're not parking gas turbines next to a data center and calling it a power strategy. That US tech companies reached for that first tells you everything. Speed over consequences, every time. And shut up about that China still fires a lot of coal. They are phasing it out... the right way.
You know, we don't actually have to do this. The people could block the gates and prevent the delivery of the turbines. The local fire department could fail their inspection and prevent the gas from being turned on. The UPS could "mysteriously" go up in smoke during it's commissioning.
None of this is preordained. We, the phyical persons, can choose how we want to live, and what constraints the corporate persons must accept.
All this at the very low cost of: your job. You can't have an unbreachable blockade 24/7 in front of there, very soon they'd send police to disperse. Fire dept has to justify the failing, if everything up to code and the inspector make it fail it will delay for at most a week and the guy will lose his credibility. The guys delivering and installing the UPS are just trying to make a living too, destroying the thing will get them fired while only delaying a little the time getting another delivered. You are replaceable.
Do you think the rest of your life will be of very high quality in any scenario?
Yeah if you want to end up in the US penal system. I assure you, you do not. I know I don't.
Bitcoin received a lot of backslash due to its power usage and ecological footprint. It's obvious that HN users don't have the same criticisms towards AI data centers. Maybe someone with a short position will coordinate that communication soon..
Bitcoin was just purely wasting power for a token only used for crime and speculation.
It’s worth noting that the ideological drive to create more energy-neutral crypto schemes resulted in trillions of dollars in fraud and no value created. Same as it always is with climate extremism. Always a cash grab.
What is climate extremism? Because it doesn't seem like the world is actually doing much to combat climate change as we see more fires and more extreme weather while still clinging to the old fossil fuel economy as long as possible.