The AI data center narrative is the perfect storm:
First most of the data center build out is happening in areas that have had little other opportunity so local resistance is muted. Abilene, Texas is referenced and is the kind of place my grandfather would lovingly say is where you go to learn to be a “dirt farmer”
Second, every environmentalist in the US is fighting 100 different battles with the most anti-regulation, pro-energy administration in decades (ever?) and has limited bandwidth.
And third, the AI narrative around national security, longevity, and super-intelligence-enabled abundance provides massive national coverage - the implications being that AI will solve any environmental and or human economic disasters that they enable.
Using GPUs as paperweights at night doesn't seem to be an efficient usage and it comes with its own costs. "Pull from the grid" is what they try not to do.
A medium AI data center uses 1GWh.. per hour.. 10-14GWh overnight (the shoulders aren't strong solar producers). Google 1GW (India), Meta 2GW (Louisiana), Meta 1GW (Texas), Stargate 1GW (UAE), Tsukuba 1GW (Japan), Gangwon Hyperscaler 1GW (South Korea), Teesside AI 6GW (UK).
On a sunny day, that will let you to power the data center fully from solar for most of the day, and partially through the night.
Obviously you need ~1GW of gas turbine capacity as well, for the non-so-sunny-days, so the economic case for solar here is dependent on the fuel price.
The Pacific Ocean is a bit of an exageration here.
There isn't a single battery this big today, but if batteries continue following the exponential growth curve they've been on then there probably will be in the next decade or so (if not sooner).
There is, and "some batteries" is an under count, however:
- The ratio between "some water" and "the pacific ocean" is a lot higher than 5
- On an exponential growth curve, a factor of 5 isn't all that much.
This probably isn't feasible for a data center being built today (although they could build solar to at least reduce their fossil fuel power generation needs during the day). But it probably will be for data centers being built in 5-10 years time.
Basically it's important to differentiate between "we can't quite do this yet" and "we'll likely never be able to do this". And powering a data center with rewnewables+batteries is definitely in the former category.
Do you have a source for this? I constantly point out that several cloudy days in a row often happens but am rebutted with graphs of the declining costs of solar modules and batteries
Nobody is suggesting to use solar as the sole source of electricity. That would obviously be insane and then leads to silly numbers like the ones you gave.
A MWh of solar is about 30 Euro/MWh if you levelize the capital costs.
The article discusses why LCOE is not a good estimate of the costs. Yes, combining different sources lowers the total cost (by damping intermittency). For Denmark it's offshore wind and solar in a 7 to 1 ratio (plus natural gas or biomethane power plants).
> Nobody is suggesting to use solar as the sole source of electricity.
The electricity price is not decoupled from the gas price because there's not enough energy storage, wind turbines, interconnections (that is the energy grid can't function without natural gas). The cost of building (and operating) all this goes into SLCOE.
No, it means that if you expect that sticking solar panels everywhere will solve all the problems for cheap (because solar panels are cheap), then you are up for a disappointment. I am of the opinion that handling out rose-tinted glasses will not go well.
Abilene is the confluence of three major resources. Massive wind farms that stretch from horizon to horizon, natural gas wells that are constantly having to flare off excess capacity, and a direct freeway connection to both Fort Worth and Dallas. In addition, land is dirt cheap.
I suspect their biggest problem will be water availability so they will need to use some other method for cooling.
I was thinking heat pumps hooked up to radiators and/or geo grids. Right now the cheapest method is evaporators which use a ton of water. It's expensive to get rid of a couple megawatts like that. Some data centers have their own microclimates that are 10-20C hotter on average.
China invested hugely into renewable energy. Their grid is strong. Ours is falling apart and our renewable energy contracts keep getting shredded. Gas powered data centers is beyond dystopian
That doom take has a few roots in truth, but it is mostly false. Our grid for the most part is very good and improving. A lot of the gloom is it is good for today, but here is why we are trying to expand it anyway - that is the gloom itself is helpful to get the needed changes made.
There is a lot of renewable energy in the US, and more is built all the time.
It's really not "doom" it's reality. Highly recommend googling the state of the us grid. It's under funded otherwise data centers would not be running on gas. Similarly one of the biggest anticipated bottlenecks for new data centers is power right now.
Although renewables are on the rise in use, huge projects have recently been cancelled in favor for gas. Meanwhile the us is going around blowing up and encouraging other gas producing nations to blow up gas infrastructure.
If you're worried abour grid health, how is adding distributed generation colocated with new loads dystopian?
Sure, it would be nice to be able to build more transmission lines and power stations wherever it makes sense for engineering to build them in order to build a strong grid. But that's hard to do with strong private land ownership and required environmental impact reporting.
Something something texas avoiding federal electric regulation.
It's unsustainable and bad for the local environments surrounding it. How many people need to die so people can chatgpt a social media post or summarize an article? These are real unanswered questions we are facing because the US has inadequate energy infrastructure.
You ever lose power and run a generator? The stink and sound is awful. Imagine powering more houses than there are in a suburban town with gas? Imagine doing that in hundreds of thousands of locations across the us. It's a dystopian thought.
I agree it is hard to stand up an effective grid to sustain technological innovations and products. It sounds like something we needed to be ready for like other countries are. Or maybe something that should happen first in order to be responsible and get ready for the future.
Last year, China installed 1.5GW of nuclear and 300GW of renewables. China's total coal use decreased as they decommissioned and under-utilized more coal than they installed.
This same pattern is occurring everywhere, regardless of local politics or local economic system. See Texas as another example.
It's because new renewables is superior at contemporary market prices. Markets have decided. Governments have decided. Everyone has decided.
It's so boring to relitigate this constantly on HN. It's like debating whether the sky is blue. Ridiculous that this comes up so often.
However you may try to put lipstick on a pig, the core of the matter is unchanged. China's coal use decreased, and they barely install any nuclear compared to renewables.
For example China has strongly expanded in making liquid fuels from coal.
"China consumes about 380 million metric tons of coal as a feedstock for chemical and liquid fuel production, representing about 8% of the country’s total coal consumption of 4,939 million metric tons."
The amount of coal that China uses only as a feedstock for chemical and liquid fuel production is bigger than total coal consumption in Europe or total coal consumption in US.
Nuclear is useful but it's hard to see it as a panacea. Renewable energy on the other hand is hard to beat. The economics of it keeps getting better, and previous estimates of the lifecycle of things like solar was grossly misrepresented.
Cancelling wind power contracts etc was a huge mistake.
"Environmentalists" is a large, diverse group and nuclear energy has been a controversial topic splitting the group for decades.
Many environmentalists are pro-nuclear, and viewed exclusively through an environmental lens, nuclear is likely the best energy source.
Other people share the "environmentalist" label because they care about clean air, unpolluted rivers, biodiversity, climate change, etc but they oppose nuclear on unrelated grounds (eg, as part of an anti nuclear weapon proliferation agenda) or out of fear of adverse events from damage to an energy facility.
The "pro-environment but anti-nuclear" subgroup held power within the Democrat party in the US through most of the cold war era. The "pro-environment, pro-nuclear" subgroup is now the largest group within the Democrat voting base, but some of the people and all of the regulations from the 1960s-1990s are still in power.
Yes, and that dates from the era when everyone thought that a "nuclear winter" (from an exchange of nuclear bombs between the US and USSR) was a more immediate risk than a long slow problem of climate change. That's what the French government carried out a terrorist attack against Greenpeace in New Zealand for: the Rainbow Warrior was protesting against French nuclear weapons testing.
Has everyone forgotten that the current crisis kicked off over the question of exactly how much uranium of what purity Iran is allowed to have? Do people really think that every country in the world should have multiple nuclear reactors?
yes, but that is not the point. The point is that countries pursued uranium reactors as a priority to produce plutonium for weapons. It did not / does not have to be this way, and the conflation of all nuclear reactors as either being at risk of meltdown or linked to nuclear weapons has resulted in the current situation. The green movement did us all a disservice by choosing not to clarify the difference, and here we are in 2026 still burning coal, which continues to produce more radioactive waste than all the nuclear reactors ever built - and released directly into the environment!
There are many claims that Thorium fuel cycle is nuclear proliferation safe, it is safer but not perfectly safe. Uranium-233 bread during the Thorium fuel cycle can used to construct nuclear weapons
"A declassified 1966 memo from the US nuclear program stated that uranium-233 has been shown to be highly satisfactory as a weapons material, though it was only superior to plutonium in rare circumstances."
Many nuclear energy opponents require absolute nuclear proliferation safety.
I was under the impression that Carl Sagan basically made up the concept of nuclear winter because of his political alignment.
And, yes, every country should have nuclear power and nuclear deterrence. Please contrast how the US negotiated with North Korea and Libya. Or what happened to Ukraine after they denuclearized. You might be impressed at how palatable peace becomes when leadership has to consider their personal safety and not just cynical economic and political calculus in their use of force.
Please. Show me a nuke that was under 10 billion over budget and less than 20 years late.
Seriously. Don't even try to insert nuclear power into the debate before you show it can be competently built, never mind safely run and waste safely disposed. But what about the French you say! The French are discovering they forgot to set aside money for decommissioning.
If you look at how much time and money the French poured into their nuclear plant construction project (and several other pre-decommissioning issues) things look rather less rosy.
Also, it was a very high priority for France - keep in mind Napoleon's quote about the relative importance of moral and material.
More critically: The logic of "The French did {extremely complex thing} fairly well in the late 1900's, therefore we can also" is very similar to the logic of "SpaceX designs and operates reusable rockets with hundreds of launches per year and no failures, therefore we can also". NO, sorry, you are just fantasizing about rocket science somehow being easy for you. SpaceX has lots of highly motivated would-be competitors - many of them far better financed than SpaceX was - but the hard fact is that zero of those can actually do what SpaceX is doing.
The costs were not small, but probably smaller than the costs of German Energiewende. Messmer plan needed not only money, but also capable heavy industry and consistent political support.
Because it’s a dangerous technology where profit over security can have severe consequences.
Not to mention the new risks now that drone wars are a thing. All those Chemical plants are already valuable target, no need to additional nuclear ones.
>Because it’s a dangerous technology where profit over security can have severe consequences.
Yeah it's far better to have power plants kill a steady stream of people, but in a banal way that's hard to attribute, like coal power plants causing lung cancer.
No, but hydroelectric works pretty well. There's a reason they recycle aluminum, etc. (power-hungry industries—crypto-mining, ha ha) along the Columbia River, etc.
So build your data centers there. No reason to choose the least evil.
Hydro electric as a resource is probably mostly exploited - are there a lot of big hydro projects left to build? If there are they must be difficult or expensive or they would have already been built.
Even ignoring nuclear environmentalists are still the enemy here.
Green energy is diffuse. Fields of solar. Ridges dotted with turbines. Each unit needs a power cable running away from it, access roads, etc, etc. A lot of area has to be developed for a given payback (sellable power).
Environmentalists and have been instrumental in making it economically impossible to develop land cheap enough to make low value density projects like that pencil out.
The higher power cost states in the US would likely be dotted with all manner of solar infill if not for up front costs that these shortsighted and selfish people have imposed on any land alteration larger than approx SFH lot size. Farmer Johnson would love nothing more than to put up solar on that ~2ac hillside he owns but cannot farm economically. Neither he nor some 3rd party who would put up the panels will shell out half a mil for an EPA CG permit just to clear the vegetation because the panels will never pay that back in their lives.
The economics of compliance are why the only greenfield development that happens these days is value dense commercial stuff (shopping plaza, big box store, data center, office buildings etc) or dense SFH development.
The most rosy possible outlook is that we "just" wait for the hippie boomers who cooked this crap up to croak, shit can the clean water act and come up with some new way of regulating development that doesn't saddle these super low impact projects (it's hard to be lower impact than panels or turbines) with fixed-ish costs that are a non-starter except at huge scale.
But of course there are so many parties who are making rent off this broken system who won't go down without a fight.
A lot of environmentalists are just degrowthers and NIMBYs. During the Biden administration multiple groups sued to block high voltage transmission lines for clean energy.
Even if modern renewables and batteries kill the need for future nuclear power plants, that doesn't excuse the consequences of decades of burning fossil fuels because environmental groups fought against nuclear power altogether.
We had better options back then, and we chose not to implement them while slowing down efforts to improve them: nuclear reactor designs could have been standardized to lower cost, even safer and more effective reactor designs could have been pursued years earlier, etc.
The costs--and opportunity costs--of inaction during that time were massive, and we're going to be paying them for generations. Renewables have a heavier lift ahead of them as a result, with less time to build out and upgrade the grid, transition to EVs, etc. The very least we can do is acknowledge the consequences.
On the cover it's a story about the failed production of Alejandro Jodorowsky's Dune script but the deeper story was the aggregation of an unbelievably talented pool of visual artists including Jean "Moebius" Giraud (mentioned as central artist in 5th element), H.R. Giger, Chris Foss, Salvador Dali, & Dan O'Bannon.
That group would go on to centrally influence the visual style of a huge body of science fiction work including Alien, Blade Runner, Total Recall, Star Wars, The Matrix, Guardians of the Galaxy, etc etc.
The art and creativity on display in the film is absolutely sonic.
How do you expect that to work? Automatic reporting is impossible, you have to rely on individuals to arrive there, open the app and take a guess. Then by the time you see the report the line is long gone (or tripled)
> Due to the federal funding lapse, this airport has temporarily suspended wait time reporting. Allow significantly more time at security and check with your airline for flight status.
Ideally the TSA at each airport would measure it and release it. They should be measuring it anyway since they should both have efficiency targets for how much of a delay they introduce, and also so that they can show data about how much or little inconvenience they cause when DOGE finally comes to cut one of the actually utterly useless government expenditures.
Since the TSA doesn't seem to be releasing this data though, apple or google could spy on GPS and motion data for individuals to estimate when people entire the line and pass through security, and derive a better-than-nothing estimate. It does seem like the government refusing to do something, and apple/google stepping in and doing a government-like thing is a norm, so even though I'm joking I wouldn't even be that surprised.
Fascinating first principle to thinking about Kalshi 90% of it is sports betting - I assume a lot of their revenue is gambling law regulatory arbitrage atm
These platforms actually seem like a great way to run sports betting to me. Much better than what the actual companies are doing trying to write their own books and risking arbitrage.
It's a tech story wrapped in a soap opera wrapped in one of the all time finest soundtracks ever played by an incredible group of actors and written by artists - it is singular!
PS - Christopher Cantwell - one of the writers and showrunners - has written a library of wonderful comic books worth investigating
From first principles public pension funds are broken.
The "Safe Withdrawal Rate" assumed by many private individuals planning for their own retirement assumes a withdrawal rate in the 3 - 4% range based on the "trinity study" - https://en.wikipedia.org/wiki/Trinity_study
Meanwhile, American public pensions are structurally engineered around a 7%+ SWR - this was recently confirmed again by the median goal by the National Association of State Retirement Administrators.
The perpetual "under funded" nature, and all the return hunting etc in pension fund management can be explained by that disconnect.
But this then belies a very uncomfortable acknowledgement which is that we cannot afford the government workforce currently in place requiring us to either:
Individuals saving for retirement must deal with the risk that they live to an old age and their savings must last for decades. Pension plans have a higher safe withdrawal rate, because people on the average have average lifespans. When a plan member dies early, their remaining contributions can be used (partially or in full) to fund other members' pensions.
But mortality credits (pooling) don't solve the math of the discount rate - they add 100 - 150 basis points of reduction so retarget to 5.5% vs 4% if generous
So they are still structurally designed where they HAVE to allocate towards risk to meet their targets which is at core of issue
It feels like you're missing the whole point of the whole article, which is that treating public pensions like a bunch of 401ks misses the opportunity to invest all that money in something that benefits the retirees collectively. I'd rather retire on 4% from a bond to improve the school my grandkids go to than 5.5% from a PE firm that intends to "more efficiently manage" the retirement home I live in.
Only taxpayer funded defined benefit pension plans get to use 7%+. Because they have the power to use future taxpayers’ money to pay for underfunding/underperformance/corruption from the past. And obviously, politicians that would choose to increase taxes today for something that could but punted to the future would lose elections.
The Pension Protection Act of 2006 mandates that non taxpayer funded defined benefit pension plans use discount rates from high grade corporate bond yield curves, which are much lower.
Yet that is the modus operandi of governments worldwide that need to secure the votes of the old. There is no reason the US explicit has different rules for taxpayer funded and non taxpayer funded defined benefit pensions. They are the same liabilities, the same cash flows, the same probabilities, just different political entities on the hook.
Even some non taxpayer funded defined benefit pension plans are more privy than others, depending on the political power of their beneficiaries.
Yup, this is the real problem. What he's talking about isn't the real problem. The real problem is we have gone to a "model" of fund less, pretend we can make it up by increasing risk. And rather than face a big problem it gets swept under the rug until it can't be any more.
We also have the problem that paying pensions becomes somebody else's problem. Thus in contract negotiations pushing benefits into pensions rather than current wages becomes attractive. You can agree to more without breaking the current budget.
The old rules worked better because limiting what pension funds could be in also limited the shenanigans that were possible.
Well we tried the "burn it all down" approach with DOGE and the BBB and we got... -2 trillion in savings? Really? Wow, okay.
The truth is nobody wants to solve the deficit. It's a self eating beast at this point and simply cutting funding for a bunch of shit won't solve it. A lot of these things are fundamentally the federal governments concern, whether people admit it or not.
What does it mean for something to be broken from first principles? I would expect some that just cannot work on a fundamental level, like faster-than-light travel or a lightbulb that powers it’s own via solar panel.
3% vs 7% doesn’t seem broken on principle, just, a tuning parameter is off.
> But this then belies a very uncomfortable acknowledgement which is that we cannot afford the government workforce currently in place
It's not just the government. It's all of the other stuff seniors buy. It used to be that you just kind of stuck around and retired in the area where you had worked. You had paid off that house, so you retired in it. Maybe you went to the Shriners' hall and played bingo with a core group of friends until you couldn't anymore. Then you moved into a retirement home and they found activities for you to do there until Father Time came to collect his due. Maybe you spoiled yourself with a Buick or Lincoln sometime between getting your gold watch from the plant manager and croaking.
Now, that's not enough. We need entire retirement communities hundreds of miles away in warmer climes where they can play golf several months out of the year. We need cruises and travel packages. We need cosmetic procedures to look younger. We need more advanced surgeries that extend life, though not participation in the workforce. And of course, now that Lincoln is a Mercedes.
And that's great, because we've told everyone that they deserve it after a long, hard career. There's only one problem: we never addressed where that retirement income was actually coming from. They're coming from that 7% SWR, which must be funded somehow. Otherwise the retirees might have to stay in-town, be cold during the winter, and provide childcare for the grandkids because preschool now costs as much as a year of college tuition. And that makes them cranky and they start calling investment advisors and politicians demanding answers.
"Teleoperation makes this even stranger. . . There are people in one country sitting at desks, driving forklifts in another country . . . It feels like immigration without immigrants."
This is a fascinating point - if Neo / Tesla deliver a teleoperated hybrid at their <$30k price point the low-skill US labor force is going to be significantly disrupted on a shorter timeline than I would have previously estimated.
These are being pitched as "home robots" but clearly corporations will go all in - 24/7 operation (with multiple remote operators), no labor law / healthcare / pensions, spin up / down at will.
I'm not so sure. The tech to do this has been around for ages, and it still hasn't happened. So I'm thinking there's something else preventing companies from going this direction.
My uneducated guess is that if a remote operator has a bad day, there is nothing stopping them from doing damage on potentially sensitive and expensive assets and then disappearing in a country with lax enforcement.
Also, after a certain point, you need to deal with the angry, hungry mob right outside your factory.
I disagree. The advent of high-speed internet in poorer countries is only something that has taken place in the past 10 years and it's only going to get worse.
This is almost like a second stage of the manufacturing revolution that happened with the advent of computer numerical control and generally the digitization of the so-called engineering "stack"
Right now, the pipeline is looking super gross for anyone apart from hyper-capitalists, but then again what meat is left to pick off the bone? Pretty much everything is overseas nowadays anyways
- Engineering Prototype is done domestically
- Verification is offshored to low-rate engineers
- Final CAD drawings are offshored
- Final Assembly instructions are offshored
- Production line is designed domestically
- Production line is assembled offshore
- Production is manufactured and shipped domestically at the lowest rate
- Said product is handled in a warehouse that is operated by offshored teleoperators
- Customer support for the end user is offshored
- RMA process is non-existent therefore replacement is offshored
>My uneducated guess is that if a remote operator has a bad day, there is nothing stopping them from doing damage on potentially sensitive and expensive assets and then disappearing in a country with lax enforcement.
Can't people already do massive amounts of damage to a company truck/van by driving it into the water, or dumping gas on it and igniting it? Doing it remotely only makes marginally easier, but most people won't do it because they don't want to be on the lam just to send an anti-capitalist message.
I'm saying if your remote operator in some southern hemisphere country causes damages, you might have less recourse for punishment than if the operator was local and under your jurisdiction.
Years ago Marvin Minsky gave a talk before 2001: A Space Odyssey played. He casually mentioned that if NASA (or was it DARPA?) had invested in tele-robotics like he insisted, your house would be cleaned by someone in Africa right now.
"When I was a kid, science fiction was off in the future somewhere. . . now it’s kind of science fact. . . AI and social media and all these influences are driving progress so quickly, at a pace that I think many of us, as human beings, are having a hard time keeping up with."
This is a quote from Alex Proyas (director of Dark City, irobot, the crow etc), he has long observed that humans are biologically under-adapted for a modernity of silicon-fueld complexity
First most of the data center build out is happening in areas that have had little other opportunity so local resistance is muted. Abilene, Texas is referenced and is the kind of place my grandfather would lovingly say is where you go to learn to be a “dirt farmer”
Second, every environmentalist in the US is fighting 100 different battles with the most anti-regulation, pro-energy administration in decades (ever?) and has limited bandwidth.
And third, the AI narrative around national security, longevity, and super-intelligence-enabled abundance provides massive national coverage - the implications being that AI will solve any environmental and or human economic disasters that they enable.
reply