I would be very skeptical about Meta on top of running any reactor. Move fast and break things doesn't work out well in that area of business. Why don't they invest money into more renewables instead, to power their data centers? That seems a safer bet and choice.
How big are the commitments here? I’m having trouble finding actual dollar amounts. Does this actually represent an infusion of money into these SMR efforts, or are these “commitments” tied to so many missable targets that it’s actually meaningless?
Oklo in particular seems to be total vaporware, I can’t find a single technical picture anywhere of anything this company’s reactor is seeking to do. They seem to raise money based on a rendering of a ski lodge.
A huge, concrete investment in TerraPower would be more interesting, but as a molten salt SMR which has never been built, this also looks extremely non-committal.
SMRs in general seem like a dead end, we’ve heard about them for decades and they don’t seem to be any closer to making nuclear power buildouts less expensive.
Everything that makes proven nuclear power plant design expensive seems to revolve around the same drivers of expense for all long-term construction: large up front capital requirements, changing regulations, failure to predict setbacks, and pervasive lawsuits. SMRs purport to tackle a couple of these (shorter-term builds, fewer setbacks), at the cost of considerable efficiency, but so far this seems like an inferior alternative to “just get better at building proven nuclear plant designs”.
China are building dozens simultaneously, and even with their questionable workers rights, safety and environmental practices, they cost $7 Billion a pop.
Cheap-er, not cheap. They’re still fundamentally massive complicated constructions. They will never be as amenable to mass production cost reductions as things like solar and battery
Can we please not have these "slightly improved language" comments? You're arguing against something I didn't say and making a meaningless nitpick on word choice.
you literally said "cheap" and the comment said "cheap-er not cheap". I think the comment is correct and you are wrong. China is building the same design again and again and again. And it's still not cheap.
tbh i don't think either the original or improved language post is presenting effectively because they both just give a conclusion without any nuance, explanation or support. "cheap" cheaper who cares? $/kwh matter. transmission costs matter.
Okla really seems like a meme stock. Their original design was rejected by the NRC, so they are very far from ever breaking ground.
I don’t understand why their valuation is so high. Why not just take all this money and build an existing, approved design?
I will be very surprised if Oklo makes it. Insiders have been selling a fair bit over the last couple years because my speculative guess is they know that they cant possible meet the expectations in the market for their product.
They essentially got a ton of traction because Altman was on the board (but since left) but most (not all) tech people don’t understand deep energy problems.
Basically it sounds like what happens in failed countries:
> “It’s not like the NRC asks for an extraordinary amount of information,” said a former nuclear official who was involved in reviewing Oklo’s failed application and spoke on the condition of anonymity to avoid jeopardizing their work in the industry. “The NRC asks three questions: What is the worst that can happen, what are the systems, structures and components in your reactor that prevents that from happening, and how do you know that?”
>“Oklo would only answer them at a very high level,” the person said. “They wanted to say nothing bad can happen to our reactor.”
>DeWitte said Oklo had planned a robust public rebuttal but claims that at the time, NRC officials “threatened us, in a retributional way, not to issue a response letter to correct the record.”
> “It’s not like the NRC asks for an extraordinary amount of information,” said a former nuclear official who was involved in reviewing Oklo’s failed application and spoke on the condition of anonymity to avoid jeopardizing their work in the industry. “The NRC asks three questions: What is the worst that can happen, what are the systems, structures and components in your reactor that prevents that from happening, and how do you know that?”
This...does not square with the successful hamstringing of the nuclear energy industry by regulation over the past several decades.
Are you saying that the NRC asks for more than that? That there was a different process in the past? The big complaint I've heard about the NRC are changes required mid-construction, which happened last in the 1980s.
In the 2000s the NRC adopted a new licensing scheme at industry urging. What "hamstringing" are you talking about?
Okla would sound a lot more reliable here if they would have fought back with lawsuits with their accusations, or if the would release the communication now that there's no chance of this supposed retribution. As it is Okla makes all the talk of "hamstringing" seem like people not doing their jobs and trying to blame others.
Without speaking to Okla specifically--I think it's completely reasonable (if not accurate or charitable) to assume they're avoiding as much compliance as possible--the simple fact is that, under the watch of the NRC, there have been a tiny number of licenses issued.
If your agency's job is to regulate something and you've done it so successfully that barely anybody has actually gotten a license--all while complaining about compliance costs--maybe you're the problem.
Had the fellow said "Oh, we have a really high bar for safety and compliance, and not everybody's able to handle that", it'd be fine. But, acting like "oh golly gee we're so easy to work with we don't ask for much" is brazen horseshit.
The data doesn't fit for NRC being the problem. Hell look at the Summer reactor that was approved alongside Vogtle in Georgia: construction failure, billions wasted, and none of it to do with the NRC.
Or any the other many many other reactors abandoned at various states of development:
There's an argument that the NRC could do things better, but placing all the well documented failures in the nuclear construction industry on the NRC doesn't make sense. Who are going to believe, the people who are always late and over budget, or the bystanders in the industry that have watched it all play out?
The expectation is that their close-to-Trump investors will push for the dismantling of the NRC, which is something Republicans wholly support, which will of course make their rejection moot.
Maybe but the underlying tech still needs to perform which, as i understand it from public docs, has not. No amount of clear runway will make up for an airplane not being able to take flight.
The NRC is now mostly composed of Trump appointees. They’ve been quietly doing that. His most recent appointee was just made chair. Expect permits for friends of the Trump family and heavy regulation for competitors.
Let's be honest it's one of the smartest and most useful place anyone could be investing - it's literally is whatever happens a way to contribute to mankind - even if it's just so that FB servers are off-grid ; it's still a huge win, I just hope Mark realize he has much more potential than what he is doing rn
The progress on 'nuclear' is so slow, that the same investment in Batterie and renewable would actually help a lot more around the globe.
We know how to build nuclear, we don't do it because its too expensive. Other forms are so far away from being useful, that the current Storage + Renewable pricing is so crazy good, that whatever you do with nuclear will just not be able to compete.
And the benefit? Every 3th world country and person can invest in small and big Storage + Renewable but they can't do the same with nuclear.
>We know how to build nuclear, we don't do it because its too expensive.
Refusing to build nuclear for decades makes it more expensive. If we start actually building reactors the cost will come down.
>the current Storage + Renewable pricing is so crazy good, that whatever you do with nuclear will just not be able to compete.
I would find this more persuasive if there were no new investment in carbon sources, but carbon sources have clearly remained competitive with batteries + solar, and global carbon emissions remain at an all time high. There's demand for baseload energy.
Building nuclear power stations includes a lot of labor-intensive hard to automate tasks like construction. Baumol's cost disease means it's getting even more expensive: rising general productivity leads to higher wages and higher costs in fields that cannot increase productivity as much as the general economic growth. That's why it's also still cheaper in countries with access to low-cost labor.
SMRs are a try to get out of it by building more but smaller reactors. The reality is however that nuclear has an issue with scaling down. Output goes down way faster than costs and most SMR designs have outputs far greater than what initially counted as an SMR.
Investment in renewable energy already greatly outpaces investment in fossil energy. The economic decision to keep using a fossil system is a different one than having to choose a new one. There's still problems that have no economically competitive renewable solution yet, but a lot of what you are seeing is inertia.
Base load electricity is simply an economic optimisation: demand is not flat, but the cheapest electricity source might only be able to create a relatively flat output. You'll need more flexible plants to cover everything above the base load. If you have cheap gas, base load does not make any sense economically.
For the last two years more than 90% of new power generation capacity added globally was renewable. Est 95% in 2025. So no, new carbon sources are not competitive.
Highly misleading stat. That's referring to capacity expansion, not new construction.
Prior energy assets go offline and are replaced each year. The report you cite is discounting all of that, looking only at expansion above the baseline, then taking total renewable construction and calcuating renewable total construction's share of expansion. Apples to oranges.
If you look at the chart in your own link you'll see that carbon construction investment exceeds renewables still.
Chart: "Annual energy investment by selected country and region, 2015 and 2025"
I would love for what you say to be true but it just isn't, even by that agency's own stats.
It's better than carbon. And solar + battery requires more carbon to produce than nuclear energy as there's a lot of mining and physical construction involved + you must overbuild to supply power or rely on non solar sources.
All for building solar. Do not understand the constant need to denigrate nuclear in favour of carbon sources while doing so.
(If carbon sources were at zero this would be a different conversation)
Nothing inherently wrong with steam, just as there's nothing inherently wrong with spinning rust hard disks or punch cards.
We are at the end of the tech curve for steam, we have pushed it hard and made some super impressive technology, but it's not advancing anymore. Supercritical CO2 might have some advantages, or other fluids.
We have zero-carbon tech that uses non-steam principles, and is currently on a tech curve that's getting cheaper than any thermodynamic cycle. We have storage tech now which is an even bigger revolution for the grid than cheap solar, because a huge limitation of the grid has always been the inability to store and buffer energy.
I still have pinning rust disks, but only because they are cheap. If SSDs were cheaper, then we would see a massive switch.
(BTW denigrating steam also denigrates all fossil fuel electricity sources, because they use the same mechanism, except for some natural gas turbines)
> Nuclear is expensive even after the reactor is build.
Solar panels and wind turbines need maintenance too. And they have much shorter operational lives than nuclear power plants, meaning they'll need to be expensively replaced much more frequently.
> And I wouldn’t call it progress to still rely on steam machines for energy
Could you please explain your objection to steam-based power? Is it purely aesthetic, or is there some inherent downside to steam turbines that I'm not aware of? Also, concentrated solar power systems that concentrate sunlight and use it to boil steam[1] are significantly more efficient than direct photovoltaics.
There's the "we forgot" hypothesis, but I think a more realistic hypothesis is the "we got too rich" hypothesis.
Construction productivity has stayed stagnant for more than half a century, while manufacturing productivity has sky rocketed and made us all fabulously wealthy compared to when the first nuclear reactors were built half a century ago.
I don't trust China's public cost numbers as much as I trust their actual capital allocation on the grid. And I will trust GE's numbers once they have actually produced something at those numbers, as pre-build cost estimates for nuclear are not believable due to their extensive track record.
Since you say ALARA made things expensive, maybe you can tell me how you foresee cheaper designs without it?
There's a lot of talk and some very shady science about getting rid of ALARA but nobody says what will change on the build that is causing the cost. Meanwhile China has adopted the same designs as in the West, without abandoning ALARA.
Those who advocate for changing ALARA see to be mostly trying to shift the Overton window on the public opinion of radiation rather than trying to pursue engineering and cost goals. I hope I am wrong on that!
Nuclear + Batteries could be nice too because the reactors will be always working at optimal rate without having to start/stop them to adapt to demand and let the storage manage peaks and lows. So investment in one domain can help the other too.
The promise of small nuclear reactors, modular reactors, thorium or whatever else has really failed to materialize at the same time that solar and battery has just leapfrogged the entire field. Nuclear has some big advantages, but it's still mired in humongous upfront costs and the intractable issue of nuclear waste. And I think we're also about to see an explosion in enhanced geothermal. The good kind of explosion.
Mark is doing much more than should already, and not the good kind, no at all!
This have a slight potential of becomeing a good one, if we only dream good things. Very limited details here, pure corporate self paise dominantly, can become anything. Another bad for example.
They're just purchasing power from existing nuke plants. All this is doing is making a formerly publicly available resource private. This will drive up energy prices for everyone else. These datacenters need to build their own generation. When people talk about how important it is that the US lead innovation in AI what they're really saying is how important it is for their quarterly results
> Our agreement with TerraPower will provide funding that supports the development of two new Natrium® units capable of generating up to 690 MW of firm power with delivery as early as 2032.
> Our partnership with Oklo helps advance the development of entirely new nuclear energy in Pike County, Ohio. This advanced nuclear technology campus — which may come online as early as 2030 — is poised to add up to 1.2 GW of clean baseload power directly into the PJM market and support our operations in the region.
It seems like they are definitely building a new plant in Ohio. I'm not sure exactly what is happening with TerraPower but it seems like an expansion rather than "purchasing power from existing nuke plants".
If history repeats itself ... tax payers will be fitting the bill. Ohio has shown to be corrupt when it comes to their Nuclear infrastructure. [0] High confident that politicians are lining up behind the scenes to get their slice of the pie.
The weasel wording is strong here. That's like me saying that buying a hamburger will help advance the science of hamburger-making. I'm just trading money for hamburgers. They're trying to put a shiny coat of paint on the ugly fact that they're buying up MWh, reducing the supply of existing power for the rest of us, and burning it to desperately try to convince investors that AGI is right around the corner so that the circular funding musical chairs doesn't stop.
We got hosed when they stole our content to make chatbots. We get hosed when they build datacenters with massive tax handouts and use our cheap power to produce nothing, and we'll get hosed when the house of cards ultimately collapses and the government bails them out. The game is rigged. At least when you go to the casino everyone acknowledges that the house always wins.
Well in a way they are building their own generation by paying elevated prices for nuclear to keep it running, as most nuclear will be shutting off pretty soon due to cheaper alternatives.
Electricity generation is getting cheaper all the time, transmission and generation are staying the same or getting more expensive. Nuclear plants get more expensive the more of them we build, but for already paid-off nuclear reactors there's a sweet spot of cheap operations and no capital costs before maintenance climbs on the very old reactors.
Meta paying for all that very expensive maintenance is not a bad deal for others, unless market structure is such that the price for entire market is set by this high marginal generation from uneconomic aged plants.
> Electricity generation is getting cheaper all the time, transmission and generation are staying the same or getting more expensive
I'm not sure what you're trying to say here, since you claim that generation is getting cheaper, staying flat, and getting more expensive all in a single sentence.
But I can tell you my energy bill hasn't gone down a single time in my entire life. In fact, it goes up every year. Getting more (clean!) supply online seems like a good idea, but then we all end up paying down that new plant's capital debt for decades anyway. Having a company such as Facebook take that hit is probably the best outcome for most.
Oops, that's a typo, should be transmission and *distrbution
Electricity costs have two components: "generation" to put power on the grid, and then the "transmission & distribution" costs which pay for the grid. You can likely see the costs split out on your bill, and the EIA tracks these costs.
Generation costs are falling, because of new technology like solar and wind and newer combined cycles natural gas turbines. However the grid itself is a bigger part of most people's bill than the generation of electricity.
Most utilities have guaranteed rates of profit on transmission and distribution costs, regulated only by PUCs. T&D tech isn't getting cheaper like solar and storage and wind are, either, so that T&D cost is likely to become and ever greater part of electricity bills, even if the PUCs are doing their job.
Generation in many places is disconnected from the grid, and when somebody makes a bad investment in a gas turbine, then the investor pays for that rather than the ratepayers. Look at Texas, for example, where even being at the center of the cheapest natural gas in a country with exceptionally cheap natural gas, solar and battery deployments hugely outpace new natural gas. That's because investors bear the risk of bad decisions rather than rate payers.
In places that let utilties gamble their ratepayers money, and where the utilities only answer to a PUC that gets effectively zero media coverage, there is a massive amount of corruption and grift and fleecing of rate payers.
A MW of nuke capacity is not replaced by a MW of solar or wind. New generation is much cheaper, but only because we are neglecting the parts of it that are hard and expensive - storage and transmission. Renewables without those things are worse than nuke - they are undispatchable like nuke and they are uncontrollably variable. We should build more renewables, but it is essential that we either tolerate intermittent system outages or massively improve transmission and storage, the generation is the least important part right now .
> New generation is much cheaper, but only because we are neglecting the parts of it that are hard and expensive - storage and transmission.
That's not correct, including storage with solar is still cheaper than nuclear. That's not measuring the cost by MW or GW, it's by measuring the cost of kWh, or the levelized avoided cost of energy, or the whatever metric you want.
And solar has the benefit of being able to avoid a good chunk of transmission by placing it at the site of use, so including transmission costs can only be to the benefit of solar.
You could just as accurately sum it up by saying they would like to tie up nearly 6.6 GW, otherwise they wouldn't be making quite as large a deal. They wouldn't be doing it if they didn't have a financial technique to afford it, and it's still taken a while to make the commitment.
What about less-well-heeled consumers who would be better served if the effect of increased demand were not in position to put upward pressure on overall rates?
To the extent that new debt comes into the mix, that's just an additional burden that wasn't there before and this is a very sizable investment at this scale. So the compounding cost will have to be borne for longer than average if nothing else.
Naturally some can afford it easily and others not at all.
I don't really get the antagonism with these ersatz concerns. when FB builds its own datacenters, or it's own chips & racks, or it's own algorithms absolutely no one is saying "well there's no profit motive to build a completely custom server chassis" or "oh no, theyre taking publicly available math and making it private"
Thank you for stating things so plainly, it's sorely needed on this site. The idea that success for big tech means a better society for workers or citizens is laughable and should soundly be rejected. They need to be broken apart yesterday.
Alternatively a more optimistic and high potential future is more plentiful and cheaper more reliable power and transmission is a huge win for society. So Id say its a perspective issue.
We in Ohio don't need more nuclear. The costs for maintaining what we have is already falling behind solar and wind (including batteries). Then there is the ecological costs that rarely get factored in.
And all these new datacenters are pushing up our electric bills. Maybe this deal could be competitive long term with newer reactor designs and if they are competently executed, but I'm very skeptical.
Well, because it means that other energy generation sources like oil, gas, and coal aren't being used there instead. Since they cause far, far more harm than nuclear waste does, it's a net win.
the waste isn't a win, of course, but is a downside of a tradeoff that is massively weighted to the upsides for society -- that is (otherwise) completely clean always-on high capacity energy production.
We understand very well how to safely handle nuclear waste and make it a very (very) low risk downside.
looking backwards in the supply chain for other externalities is a good point, but I'm not sure any energy production method is exempt from this?
Also, by the way, my perspective isn't about nuclear Vs X (wind turbines etc) - I like all the ones that are net clean and useful in different circumstances as part of a mix.
I'm just addressing the narrower point about whether nuclear per se is a net benefit for society, which I believe it is, massively.
I wish that Meta would pay for the extension of Diablo Canyon in California. They have had to jack up already sky-high electricity rates to keep it going, after deciding nearly a decade ago that it would be uneconomical to try to extend its lifetime.
Meta's nuclear intention is a perfect example of how tech is willing to pay far more for energy than other customers, and how it's driving up everybody's costs because we are all paying for that increase at elevated prices.
Nuclear is extremely expensive, higher than geothermal, renewables backed by storage, and natural gas. Nuclear is good for virtue signaling in some communities, but from the technological and economical perspectives, nuclear is very undesirable and unattractive. It's only social factors that keep alive the idea of new nuclear in advanced Western economies, not hard nosed analysis.
Here's a new preprint from Germans showing that even for Europe, a continent with very poor solar resources for many countries, new baseload is not the most economical route:
Geothermal energy is highly location-specific (it's great if you have a handy volcano nearby, of course), storage cannot meaningfully "back" intermittent renewables because it's only good for a few hours load from the grid (aside from pumped hydro which is effectively built out), natural gas peaker plants are very expensive and increase CO2 emissions. There isn't much of an alternative to nuclear.
Enhanced geothermal using fracking techniques developed in the past decades is being deployed now, and has very limited restrictions on where it can go.
> storage cannot meaningfully "back" intermittent renewables because it's only good for a few hours load from the grid
That's trivially untrue. If you can build a battery for four hours duration, then of course you can build another to get to 8, or however many you want. Batteries are being added to the grid at a huge rate.
> natural gas peaker plants are very expensive and increase CO2 emissions. There isn't much of an alternative to nuclear.
Gas peakers are about the same cost as nuclear. We will have a ton of gas peakers hanging around in the coming decades, and they will be used less and less as we get more batteries on the grid. Already, batteries eat peakers' lunch economically.
In the most optimistic grid modeling scenarios, nuclear can play a 5% or 10% role in a fully decarbonized grid. If you go full nuclear, then you also need a ton of batteries. And if nuclear was cheap, I would advocate going for a ton of nuclear plus batteries. But nuclear is super expensive, and doesn't scale fast enough to meet our needs.
We are currently at 20% nuclear power in the US and have a rapidly aging fleet. Even if we had investors who wanted to spend the $500B it would take to keep 5% of our grid as nuclear, it's unlikely we'd be able to build those 50 1GW reactors over the course of the next 20 years. Scaling SMRs seems even less likely.
I really hope I'm wrong and the SMRs somehow materialize and are cheap, but none of the startups are acting like the have anything real or the chops to scale. A new reactor getting built by 2032, as suggested here? Pumping out an SMR in 6 years, when design isn't even finalized, the company hasn't shown progress since abandoning an NRC application, does not seem plausible.
Meta's nuclear intention is a perfect example of how tech is willing to pay far more for energy than other customers
I believe their plan is to convert that energy into revenue at a rate that exceeds the amortized cost of generating the energy. It's not a social good project even though you interpret the cost outlay as such?
Nuclear isn't extremely expensive; in China it costs around $2/watt of power (compared to up to $14/watt in the US). It's just expensive in America because America's shit at building nuclear power plants efficiently.
Nuclear compliance and certification is extremely expensive. The actual construction and maintenance costs are fairly trivial.
The largest cost associated with a new nuclear plant are the interest payments given that a plant may need to spend 10+ years sitting idle before it can be activated.
> The actual construction and maintenance costs are fairly trivial.
That is not at all what I have seen, the costs tend to be from absolutely massive infrastructure needed to last a long time in harsh conditions that are difficult to repair.
Those seem more like fundamental engineering requirements.
Across four different regulatory structures: France, Finland, the UK, and the US, modern nuclear has proven to be excessively expensive and require massive amounts of high skilled labor. In the past century, high skilled labor was cheaper, but these days we need to pay welders and other construction workers higher wages because they have high productivity alternative jobs that pay better than in 1970.
Those high interest payments for 10+ years are also because EPC promises to build the design within 5-7 years then takes 2-3x the time. At Vogtle the fuckups both on design meant that many plans were "unconstructable" and then construction proceeded anyway with whatever they could wing together then they had to go back and make sure that whatever the bell was built still met the design.
What, we should have a wild west where everyone can set up their own nuclear power station without any compliance or certification? If not then these are part of the build cost... it's like saying we shouldn't include testing as part of the cost of building software.
You can say the same for cars, houses, appliances, medical devices, elevators, stairs, disabled access, etc etc.
So, what exactly is your point? Yes, everything would be "much cheaper" if nobody had to pay as much attention to most details any more. Everything would also be much much more expensive for everybody else and longer term, or not work at all or reliably or safely.
It's a question of magnitude. Do you think that over-regulation of specific technologies is possible?
If the price of building stairs was growing each year in only the west to the point were we were opening one staircase 5 every years, it might be worth to ask some companies why. If they all say "the last guy who built stairs got bogged down for 25 years trying to meet all the safety standards". It might be time to relax some of them.
They're not winding down per say, nuclear has stayed steady, renewables simply exploded. Nuclear understandably more prone to schedule delays, but now it's mostly execution delays in months vs regulatory delays in years. They're still on trend with nuclear year plans. Solar simply scales much faster because all the displaced construction worker from real estate slow down can slap panels but not reactors.
Curious how this new design addresses the biggest safety challenge of nuclear reactors (the issue that was the root cause of the Fukushima accident and an indirect cause of Chernobyl): how do we ensure that the nuclear core temperature remains controlled during exceptional events (e.g., earthquakes, structural failures) when the reactor must shut down abruptly?
Chernobyl and Fukushima were different accidents and causes. Chernobyl was a systemic failure of the soviet system. Fukushima was a wild edge case that an earthquake and tsunami drained the coolant.
The truth is, all reactors ever built were considered safe at their time with whatever definition of safe. No one builds unsafe reactors. Yet they turned out not to be safe.
Nothing is ever perfectly safe and a lack of perfect absolute safety is not a valid objection. All sources of power have associated risks, even renewables. Wind power has 0.04 deaths per terawatt hour and solar has 0.02 [1]. Nuclear power has 0.03 deaths per terawatt hour (safer than wind), and it's worth noting that almost all of those are from Chernobyl, which was considered unsafe even at the time (they knew about the positive void coefficient). I'm not arguing that nuclear power is perfect, mainly because it isn't. But it's not like all other sources of power are idyllic havens of safety. There are always tradeoffs.
But the bottom line is that renewable costs are trending down, hard and fast, battery tech is just getting started, and development time for wind and solar is comparatively fast.
Future nuke costs at this point are speculative, development time is very slow, and even if new reactors were commissioned tomorrow, by the time they came online it's very, very likely solar and wind + storage would make them uneconomic.
IMO the attachment to nukes is completely irrational. There are obvious economic downsides, no obvious economic benefits - and that's just the money side.
Thanks for the reply! I think you're arguing with the wrong person in the second half, though. I agree that renewables could potentially be more economically viable than nuclear power[1]. My reply was disputing the "people can die from nuclear therefore we should never use nuclear" argument, not arguing about economic viability. Also I think that broadly claiming that your opposition is "completely irrational" is not a very tactical rhetorical move.
[1]: although since you're basing your claims on the speculative future state of solar technology 10 years in the future, I don't see why the same shouldn't apply to the speculative future state of nuclear power, but that's besides the point
Apart from the deaths from workers falling off the roof or from wind turbine towers (though these might be the only type of deaths included in these figures):
If mining deaths are included, coal, oil, gas and uranium probably do not look favorable at all, but renewables aren't perfectly safe either: there was a bridge collapse at a copper/cobalt mine in Congo two months ago that killed 32. Solar and wind use more copper per energy unit than other technologies, and solar and wind indirectly require battery technology. Lithium batteries contain lithium and cobalt. (Lithium mining seems relative safe, but 70% of cobalt is mined in Congo, which is known for artisanal mining, and the above-mentioned accident indeed seemed to happen at such a mine.) Wind, especially off-shore wind uses more concrete and steel than other power generation technologies (hydro seems like it'd use a lot too?), which could be explored too. (Course, these metals are recyclable, so you only mine them once.)
Accidents, mainly. Solar panels and wind turbines produce far less energy per module than nuclear, so you have to build much more of them. If you build enough of something, the odds that everything goes perfect every single time are quite low.
Constrained edge cases are fine. Particularly when contrasted with coal and natural gas, which are, in practice, what everyone is competing against in America.
The edge case was predicted, but market and political forces chose to ignore it. The GE Type II reactor had known issues. [0]
I should add that I am not strictly anti-nuclear, and it is super interesting that some of the largest funders of anti-nuclear propaganda have been actors from the fossil fuel industry. [1]
Wild edge cases are to be expected when you do things at scale. If you build 20 buildings in different regions, at least one of them will likely face a once-in-1000-years natural disaster. And it's difficult to estimate how bad that particular kind of once-in-1000-years event can be, because you probably only have a century or two of reliable data.
Let’s not forget that you don’t have to be a socialist/communist nation to decide you want to do the cheaper thing. Without robust regulations I guarantee you a Chernobyl-like disaster could easily happen in the US because of less scrupulous companies cutting corners and choosing the cheaper path. With Chernobyl it was the government instead of a private company.
We can talk all day about how the system incentivized people playing CYA rather than actually trying to solve the problem (true and fair critiques), but when it comes down to it, this happened because the cheaper option was chosen and potential issues were overlooked. That transcends political systems.
I personally find highly hypothetical situations impossible to guarantee but I'm glad you have such a high degree of self certainty for a plausible scenario you have decided to give certain results to.
You should really consider educating yourself on the Chernobyl reactor melt down (read a book or two) to understand the level of calamity inflicted by the communist system. Stop trying to make it sound like that could happen anywhere because the pressures of capitalism could cause the same results. Its pretty eye opening how insane the chernobyl situation was.
The US had Three Mile Island. Japan had Fukushima.
One of the biggest arguments against nuclear is that reactors are insanely complex. Beyond a certain level of complexity, safety and predictability become impossible even with perfect management - which certainly doesn't exist in the nuclear industry.
This is especially true of any nuke system which needs external cooling, because stable water levels aren't a given any more because of climate change. Between floods, droughts, and storm surges, the environment is part of the system - something Fukushima discovered to its cost.
I am actually very familiar with the history of Chernobyl and the meltdown. What I’m saying is human greed and short sightedness do not suddenly go away because a nation decides on a different political/economic system. The implication that it only happened because it was the Soviet Union is what I’m taking issue with because it absolutely could happen in the US without proper guardrails. All it takes is one bad company cutting the wrong corner or firing the one person who spoke out. It’s very easy to see no society is immune to this.
I am not defending the Soviet Union or any of the decisions made during Chernobyl. So you should redirect your indignation/condescension.
You don't need to prevent it. You just need to prevent a catastrophe and even Fukushima did it relatively well - nobody died or will die from radiation. Current benchmark for (future) gen4 designs is having consequences limited to the area of the plant, think of 3MI but as worst case. But imo it's still an overkill, nuclear is one of the safest sources in terms of human deaths/kwh and the stat only gets better with gen3/3+
> You just need to prevent a catastrophe and even Fukushima did it relatively well - nobody died or will die from radiation.
“As of 2020, the total number of cancer and leukemia instances has risen to six cases according to the Tokyo Electric Power Company (TEPCO).[5] In 2018 one worker died from lung cancer as a result from radiation exposure.”
These are small numbers compared to the number that died due to the tsunami and the massive evacuation (to avoid radiation injuries). The frustrating bit is that they could have avoided it all.
"
The workers' compensation claims that have been recognized by labor authorities include six cases of workers who developed cancer or leukemia due to radiation exposure
"
So compensation has been requested for cancers, of which one death has been reported.
I remember when terrapower was supposed to build a kind of reactor which would use depleted/unenriched uranium which was basically a hole in the ground slowly burning up, but now it seems they're doing a molten sodium "classic" reactor, is that the same thing and I am confused, or did they change their strategy?
China is the world's largest electricity producer and installs a lot of generating capacity of all types. For example, China has 29 nuclear reactors with 31 GW of capacity currently under construction:
Which leads to a shrinking nuclear share in their grid. It peaked at 4.6% in 2021, now down to 4.3%.
Compared to their renewable buildout the nuclear scheme is a token gesture to keep a nuclear industry alive if it would somehow end up delivering cheap electricity. And of course to enable their military ambitions.
The nuclear share dropping is a very clear signal about a lack of investment. Shows that nuclear energy is no longer cost competitive, even in a "low regulation" environment.
It shows that strategic investment matters and people are looking at more than a single cost metric. Nuclear is behind today, but that doesn't hold a promise it will remain true into the future unless you stop investing now.
One armed bandit says explore as well as exploit. This delta you cited indicates the pendulum currently is more exploit than explore, but its not a static equation.
chinese nuclear is extremely cost competitive at 2.5bn/unit. They have other reasons, one being the ban on inland expansion fearing of messing up with 2 major rivers that feed the country. Current chinese units are basically borrowed and improved western designs, cap is basically vogtle's ap1000, hualong is a frankenstein of several western designs.
TBH this part seems key, even PRC couldn't operate full western designs reliant on western industrial capacity economically, part of it was simple incompetence of western supply chains (business closures / regulatory drama / sanctions). Nuclear seems viable once you strip out a lot of the politics that makes them uneconomical, hence PRC had to indigenize the designs since once western supply chains enter picture, the schedule goes out the window.
Two years ago we were installing 1/10th of Chinese solar today?
Where are we at today? Can we catch up under this administration?
Where do we compare on a nuclear basis? I know my state installed nuclear reactors recently, but I'm not aware of any other build outs.
In a war game scenario, China is probably more concerned about losing access to oil and natural gas than we are. Not that we shouldn't be building this stuff quickly either.
There are multiple measures, as generating technologies are complex. "Nameplate capacity" (given above) is one, "capacity factor", which is (roughly) the time-averaged output is another, and for solar averages about 20%, though that can vary greatly by facility and location.
Nuclear has one of the highest capacity factors (90% or greater), whilst natural gas turbines amongst the lowest (<10% per the link below). This relates not only to the reliability of the technologies, but how they are employed. Nuclear power plants cannot be easily ramped up or down in output, and are best operated at continuous ("base load") output, whilst gas-turbine "peaking stations" can be spun up on a few minutes' notice to provide as-needed power. Wind and solar are dependent on available generating capability, though this tends to be fairly predictable over large areas and longer time periods. Storage capability and/or dispatchable load make managing these sources more viable, however.
It's close enough to how it's measured. China's terawatt of solar power capacity isn't producing 9000 terawatt hours in a year. Their total electricity use is 9000 terawatt hours.
It is how individual power generation projects and measured though. If you install a GW of solar generation, it means you installed solar panels capable of generating 1 GW peak. If you install a 1 GW of coal generation, then same thing. If you install 1 GW peaker gas plants etc.
The coal plant will have a capacity factor of 80% though. Solar will be 10 to 20%. And the gas plant could be very low due to usage intent.
Battery projects are the same (since they're reported as generators). Whatever nameplate capacity...for about 4 hours only.
In absolute terms, China installs about as much nuclear as the US does solar. So I can only assume you agree with the statement "the US is bringing on an insane amount of solar energy"? Because, once again in absolute terms, the US's solar buildout is trounced by China's. The US is losing the energy race, and nuclear isn't going to save it. The US will run out of fissile material before China runs out of sunlight.
China is building a tiny amount of nuclear in comparison to their wind, solar, storage, and HVDC builds. Only something like 50-100GW over thw coming decades. The quantity being built only makes sense as a strategic hedge, not as a primary strategy.
Renewables crash the money making potential of nuclear power. Why should someone buy ~18-24 cents/kWh new built nuclear power excluding backup, transmission costs, taxes, final waste deposit etc. when cheap renewables deliver?
China is barely building nuclear power, in terms of their grid size. It peaked at 4.6% in 2021, now down to 4.3%.
Compared to their renewable buildout the nuclear scheme is a token gesture to keep a nuclear industry alive if it would somehow end up delivering cheap electricity.
nuclear provides for about 4-5ct/kwh if built cheap, everything included, looking at swiss data. Chinese units are built for 2.5bn/unit, so probably even cheaper than that.
But yes, china is far from what france or sweden did with nuclear per capita
Again they aren't the same product. Everyone always thinks power is only about $/kwh especially in hackernews. That is a strong proponent of the product but most definitely not all of it. Solar just does not work for large scale industrial uses cases (99.99% uptime). Even with massive energy storage to try and cover the edges. Its a great combo but not comparable.
Renewables and battery storage energy are unstoppable. Why take nuclear risk when you can get more than enough from solar, wind, and geothermal coupled with battery storage?
China is a country with over a billion people, Meta is a private company with under 100k employees, it doesn't really make sense to compare the power output of their investments.
Meta (seemingly) bought the local race track in Beaver PA for a large data center. We were sad to see a world class racing facility go away as it finally started to hit its stride. But the offer was too hard for ownership to pass up
Racetrack memory is an emerging alternative that might obviate the current DRAM shortage especially for datacenter use, so in a way this makes a whole lot of sense.
I'm kind of surprised there's so many nuclear power companies. I remember Google signed a deal a year or so ago [0] with a different one (Kairos). I wonder why they're not in this deal? Is it all connections and funny money? Or do any of these have real shots at making something useful?
I was thinking how the dotcom bust left a lot of dark fiber infrastructure which helped the internet take off after that. It would be great if the upcoming AI bust (if it happens) leaves a bunch of power generation and new nuclear tech behind.
The dark fiber analogy doesn't quite work though, because unused fiber just sits there harmlessly. Nuclear plants have massive decommissioning costs ($280M-600M+ per plant), ongoing waste storage requirements, and if the companies go bust, taxpayers are likely on the hook. The EU alone is underfunded by €118 billion on decommissioning. Meanwhile batteries and solar keep getting cheaper every year without the “who pays to clean this up in 50 years” problem. Seems like a very roundabout way to hope for public benefit.
with small companies it's like with startups - some might fail so maybe it's better to have more players. I still think this is strange - if they wanted nuclear - could have approached GE for some BWRX of ABWR or W-house & KHNP for some nice AP1000/alike units
Right. Meta wants big enough plants that an AP1000 or two would be the right size. They're known to work. There are four in operation, two in the US, and another dozen or so under construction.
Most of the small nuclear reactor startups hand-wave the failure modes and argue that they don't need the hulking big expensive containment building. NuScale claimed that.
They wanted multiple reactors sharing the same cooling pool. If they ever had a leak, the whole set of reactors would be contaminated, even without a meltdown.
If we look at the big reactor accidents so far, there's Chernobyl, with no containment building. There's Fukushima, with too small a containment unable to contain the pressure. And there's Three Mile Island, where a large, strong containment building contained a meltdown. Three Mile Island was an expensive disaster, but not hazardous outside the plant. That's the failure mode you want.
We might be better off at developing better techniques for welding thick sections to make hulking big, strong containment vessels. There's been progress with robotic welding of thick sections.[1]
> surprised there's so many nuclear power companies
This is actually American capitalism working at its finest.
Have you seen a video of a slime mold "solving" a maze? It reaches out in every direction with thin tendrils until it makes contact. (Then the game shifts.)
We have a sense, like a slime mold picking up on the "scent" of food, that there is energy. But there are lots of good hypotheses for how we get there. So we try them. Not exhaustively. But multiply. When someone demonstrates they've got it, the game will shift to consolidation and scaling.
This may just be me... but is anyone else in their head saying: "whats the catch?"
~~Meta~~ Facebook has made their money by de-stabilizing people's emotional state to keep them engaged and buying stuff via ads. I'm having a tough time connecting the dots between that and nuclear power.
I get that and scanned the article. they are currently, and have been, working on "AI" for a bit, but strategically they aren't positioned to provide a service offering right now. Even if they are going all in and they got regulatory love, it would make way more sense to go after an energy source/provider that could provide it on a faster timeframe. they'll get their energy in what? 7-10 years?
I won’t take Meta’s new ambitions serious unless Zuckerberg does something like have a “fireside chat” next to Ezra Klein moderated by Kara Swisher where he awkwardly spams the word “abundance” while staring Klein square in the brow. And then buys Vox Media.
"Infrastructure investment" is not fungible. There's either demand for this power where they're adding it, or there's not. If there is demand, then it's created by AI. If there's not (i.e. AI bubble pops), then there's excess capacity where we don't need it.
I think the arguments for this would possibly be, if AI continues to be useful (generation demand skyrockets): Meta would possibly have a positive ROI for these investments which would lead to others copying the investment strategy and building more nuclear. If that happens a large portion of AI demand would become green(-ish) energy.
If AI demand lowers (generation demand plummets): Meta would have subsidized a bunch of nuclear reactors which would likely continue to produce power for 10 years - 50 years.
A big reason I have heard for lack of nuclear build out is the lack of starting capital but after they are built they are generally stable and maintenance is predictable.
> A big reason I have heard for lack of nuclear build out is the lack of starting capital but after they are built they are generally stable and maintenance is predictable.
I have also heard this, but given Meta's announcement is mostly in funding and extending the useful lifespan, doesn't that indicate without an infusion of capital, the ongoing operations are not cost effective?
I'm glad that individuals with no democratic control are allowed to declare large swaths of society to suffering because, maybe, in the future things might get better.
FWIW, it took nearly 150 years for commoners to benefit from the industrial revolution. The idea that I must suffer and my children must suffer and their children must suffer so some future plutocrat can get a fatter nut is pathetic.
I mean, we could just give up and go back into the swamps that we crawled out off millions of years ago. But I'd rather have more clean power come online. Energy is EVERYTHING.
"Fallout risk" and "dangerous waste risk" are such non-issues that have been so absurdly elevated in the public perception that the fear of them has caused far more deaths than the risks themselves.
How exactly? Meta isn’t doing this out of generosity to society. They’ll be consuming this and vastly more energy to ultimately increase their own profits?
As technology improves, we have less and less need for nuclear. The continent with the greatest need for nuclear is Europe, and these German grid modelers have taken a look at the EU grid with the latest data and decided that additional baseload generation (like nuclear) is not required and will likely increase costs if built:
In addition to the other corrections here, I'd like to add one more remarkable fact: in 2025 the share of German electricity generated by solar increased to 18% from 14%. That's in a single year, in a country with terribly low levels of sun! Nuclear generated 5% of electricity before it was shut down, and had generated that same percentage for more than a decade (that's as far back as the chart I saw went).
It's remarkably easy to scale solar to very large amounts in short time periods. Far easier than building a new nuclear fleet.
While I'm sure that the vast majority of this energy is going straight to AI server farms, I'm also hopeful that this will renew efforts to switch to sustainable energy sources.
The entire world is switching to sustainable energy at a tremendous rate because they are the cheapest source of electricity, and storage is economical on every grid's peak price to low price swing that I have seen. In places like Pakistan solar is taking over because the grid made really poor investments in very expensive fossil fuels. In Africa, a panel and a battery is cheaper to run than a generator, which is life changing. As that scales up it will revolutionize entire economies.
In the US, the vast majority of new generation is renewables, matched with a ton of storage. There's some gas too, but it will be uneconomical to run these gas plants before their end of life:
The biggest impediment to cheap sustainable electricity is political and basic ignorance of the voting public that allows utilities to continue installing familiar, but more expensive fossil generation.
what has home solar to do with country's grid and industrial needs? No country has a VRE+BESS solution to fully sustain itself. South Australia might get close having extremely nice weather patterns but even they will periodically import from fossils neighbors.
Nuclear was dirt cheap in US but it didn't help to expand it. and now it's not that cheap since supply chain is gone
> what has home solar to do with country's grid and industrial needs?
Why are you asking this question without answering it? Unless you're assuming I said something I don't, I don't understand the relevance here.
> No country has a VRE+BESS solution to fully sustain itself.
A couple of fallacies here: non existence yet doesn't mean it's not possible, especially for a grid with equipment lifetimes measured in multiple decades. Once it makes sense to build a VRE only grid it will take several decades for it to happen.
> Nuclear was dirt cheap in US but it didn't help to expand it. and now it's not that cheap since supply chain is gone
Nuclear was completely overbuilt, based on bad over predictions for energy demand, leading to excess orders, then the only plants still being under construction being the ones that were poorly managed, leaving a distinctly bad financial taste in the utilities' mouths. They had to be bribed by state legislatures in South Carolina and Georgia to let them pre-bill ratepayers and also put ratepayers on the hook for overages on construction for new reactors, which everyone expected and which happened disastrously.
It's really unclear that nuclear is ever going to be cheap when its main costs are high skilled construction labor, concert and steel. None of those inputs are getting cheaper and the only chance at making logistics and management better seem to be small modular reactors, which are inherently cost inefficient due to losing the scale advantages of 1GW reactors.
It's no longer 1970, and the costs and technology of the modern economy are fundamentally doffeeent. We have learned so much, and I hope that we have learned tha nuclear doesn't fit with our advanced technology of the 21st century.
I despise Facebook and all that it stands for, but if the surplus value that it has extracted from humanity over the last two decades is reinvested intelligently into nuclear energy, I'm actually okay with it.
Despite the hype that you see on Twitter, the hard tech startup scene is actually incapable of large-scale engineering coordination on the level needed for a nuclear power plant, or even a gas turbine.
If any innovation on fission reactors is going to be successfully commercialized, we will need to see billions of dollars of investment over medium to long time horizons.
Of course, the millstone around the neck of nuclear power is that it's a dual-use technology. There's probably a lot more behind the scenes that's been done to stifle the industry effectively for non-proliferation reasons, but masquerading as cost, regulatory problems, environmental concerns, etc.
The biggest barrier to nuclear energy in our country is the linear no-threshold safety model. We know the model is wrong, opens up operators to infinite liability, and since adopting it the nuclear industry has effectively been dead.
Even just a basic common sense threshold would make nuclear more viable overnight.
Agreed, but I don't think that dumping ALARA will fix nuclear cost. There's still plenty of stuff in a nuclear reactor hot enough that we have to worry about keeping it from being inadvertently released.
Yes. Vistra is the largest power generator in the US [1]. Most of the deal that Meta announced is with Vistra, and the power will come from some existing facilities.
I am pro-nuclear power, but I miss the days when companies would, you know, return their profits to investors, so those investors could then invest in other companies doing different things, instead of all corporations tending towards generic everything-investment vehicles.
This is partly due to tax policy. Investors don't want this. Distributing dividends is less efficient capital distribution because those capital gains are immediately taxed rather than be deferred into a higher share price with no tax drag.
This is why every tech company does buybacks(other than meta which started a dividend when Mark got sick of high interest rates on his debt, they still mainly do buybacks though)
This is to appease the current US administration running the federal government that for some odd reasons is abhorrent to the idea of using cheap, cleaner renewable energy and battery storage to decarbonize our air and grid.
After reading through all press releases I still haven't seen any details on any money changing hands?
The previous ones from Google and Amazon at least specified that it was based on PPAs. Where all they did was bind themselves to buy X amount of power at Y cost if the company could deliver.
Taking a step back I have an incredibly hard time seeing how new built nuclear power will cope with renewables fundamentally reshaping our grids.
We're seeing the unraveling of the grid monopoly infront of our eyes [1] and renewables are set to completely crash all "baseload" markets. [2] Likely forcing them to become stranded assets.
So now we’re harnessing a the dicy and capital intensive power of the atom to create the illusion of speaking to people because we can no longer strike up a conversation with a stranger on the street. Cool, cool. Personally I’m going to stick to real people with no divergent neutron chain-reactions involved.
This is a purely political move. It will take a decade or longer if ever to ever get power from this. And yes it says "as early as 2032" but we know how that goes.
Why nuclear? Because it's cleaner than fossil fuelds but appeases the administration because it isn't wind or solar, which would immediately solve any power generation problems.
You might be tempted to say, since this always comes up, "what about base load?"
FFirst, batteries can solve that problem.
Second, you use a mix of power and when the Sun isn't out (ie night) is when power is cheaper from other sources.
Third, data centers don't really need base power at all. You just run the DC when you have power and don't when you don't. There's precedent for this. Google has a DC in Scandanavia that they shut down a few days a year when it gets too hot, otherwise it's just cooled by ocean water.
What I find most funny about all this is that all these big tech companies are kowtowing to the state in the exact same way they accuse Chinese companies of doing.
wind and solar need firming. Currently firming capacity in US is getting scarce. It can be fixed a bit with better transmission but otherwise you need to expand it. China is expanding firm power with coal, gas and nuclear. Germany will expand gas.
Bess will not solve the firming problem. And no, if you build a multibillion datacenter you want to run it around the clock as much as possible. But yes, some datacenters don't have such requirements, but here we are talking about meta
Yeah, there's a lot of stuff for nuclear. I wrote to the government to get them to remove 10 CFR Part 20 Subpart E financial requirements but obviously that hasn't happened. You can start from there and branch out.
Don't think so but after seeing electricity prices rise and even double in areas by giant data centers many areas are requiring them to provide their own power
Most of the cost of residential electricity is the distribution network, not generation. Supply/demand dynamics only really come into play during extreme scenarios like the Texas winter outage a few years ago.
The price isn’t the same, it’s more expensive, and it’s going to get worse before it gets better. This particular project is slated to take a decade — the last nuclear reactor that the US built was also slated to take a decade. It just opened in 2024, and was approved for construction in 2009. This is a long-view solution to a right-now problem.
We should have been keeping up with this infrastructure stuff all along… but I’m really not convinced all of these companies are going to be using this shit in 5 years, anyway.
I would be very skeptical about Meta on top of running any reactor. Move fast and break things doesn't work out well in that area of business. Why don't they invest money into more renewables instead, to power their data centers? That seems a safer bet and choice.
How big are the commitments here? I’m having trouble finding actual dollar amounts. Does this actually represent an infusion of money into these SMR efforts, or are these “commitments” tied to so many missable targets that it’s actually meaningless?
Oklo in particular seems to be total vaporware, I can’t find a single technical picture anywhere of anything this company’s reactor is seeking to do. They seem to raise money based on a rendering of a ski lodge.
A huge, concrete investment in TerraPower would be more interesting, but as a molten salt SMR which has never been built, this also looks extremely non-committal.
SMRs in general seem like a dead end, we’ve heard about them for decades and they don’t seem to be any closer to making nuclear power buildouts less expensive.
Everything that makes proven nuclear power plant design expensive seems to revolve around the same drivers of expense for all long-term construction: large up front capital requirements, changing regulations, failure to predict setbacks, and pervasive lawsuits. SMRs purport to tackle a couple of these (shorter-term builds, fewer setbacks), at the cost of considerable efficiency, but so far this seems like an inferior alternative to “just get better at building proven nuclear plant designs”.
The thing which can make nuclear cheap is building a large number of the same plant design.
China are building dozens simultaneously, and even with their questionable workers rights, safety and environmental practices, they cost $7 Billion a pop.
Cheap-er, not cheap. They’re still fundamentally massive complicated constructions. They will never be as amenable to mass production cost reductions as things like solar and battery
>Cheap-er, not cheap.
Can we please not have these "slightly improved language" comments? You're arguing against something I didn't say and making a meaningless nitpick on word choice.
you literally said "cheap" and the comment said "cheap-er not cheap". I think the comment is correct and you are wrong. China is building the same design again and again and again. And it's still not cheap.
tbh i don't think either the original or improved language post is presenting effectively because they both just give a conclusion without any nuance, explanation or support. "cheap" cheaper who cares? $/kwh matter. transmission costs matter.
Who doesn't say nuclear is more expensive?
The lowest LCOE for nuclear is to the right of the most expensive solar plus storage.
Okla really seems like a meme stock. Their original design was rejected by the NRC, so they are very far from ever breaking ground. I don’t understand why their valuation is so high. Why not just take all this money and build an existing, approved design?
I will be very surprised if Oklo makes it. Insiders have been selling a fair bit over the last couple years because my speculative guess is they know that they cant possible meet the expectations in the market for their product.
They essentially got a ton of traction because Altman was on the board (but since left) but most (not all) tech people don’t understand deep energy problems.
Basically it sounds like what happens in failed countries:
> “It’s not like the NRC asks for an extraordinary amount of information,” said a former nuclear official who was involved in reviewing Oklo’s failed application and spoke on the condition of anonymity to avoid jeopardizing their work in the industry. “The NRC asks three questions: What is the worst that can happen, what are the systems, structures and components in your reactor that prevents that from happening, and how do you know that?”
>“Oklo would only answer them at a very high level,” the person said. “They wanted to say nothing bad can happen to our reactor.”
>DeWitte said Oklo had planned a robust public rebuttal but claims that at the time, NRC officials “threatened us, in a retributional way, not to issue a response letter to correct the record.”
>“Well, they’re gone now,” he added.
https://www.washingtonpost.com/business/2025/11/26/nuclear-e...
> “It’s not like the NRC asks for an extraordinary amount of information,” said a former nuclear official who was involved in reviewing Oklo’s failed application and spoke on the condition of anonymity to avoid jeopardizing their work in the industry. “The NRC asks three questions: What is the worst that can happen, what are the systems, structures and components in your reactor that prevents that from happening, and how do you know that?”
This...does not square with the successful hamstringing of the nuclear energy industry by regulation over the past several decades.
Are you saying that the NRC asks for more than that? That there was a different process in the past? The big complaint I've heard about the NRC are changes required mid-construction, which happened last in the 1980s.
In the 2000s the NRC adopted a new licensing scheme at industry urging. What "hamstringing" are you talking about?
Okla would sound a lot more reliable here if they would have fought back with lawsuits with their accusations, or if the would release the communication now that there's no chance of this supposed retribution. As it is Okla makes all the talk of "hamstringing" seem like people not doing their jobs and trying to blame others.
Without speaking to Okla specifically--I think it's completely reasonable (if not accurate or charitable) to assume they're avoiding as much compliance as possible--the simple fact is that, under the watch of the NRC, there have been a tiny number of licenses issued.
From the horse's mouth in 2012, only 3 (*3!*) such licenses had been granted in 30 years ( https://www.eia.gov/todayinenergy/detail.php?id=5250# ).
If your agency's job is to regulate something and you've done it so successfully that barely anybody has actually gotten a license--all while complaining about compliance costs--maybe you're the problem.
Had the fellow said "Oh, we have a really high bar for safety and compliance, and not everybody's able to handle that", it'd be fine. But, acting like "oh golly gee we're so easy to work with we don't ask for much" is brazen horseshit.
The data doesn't fit for NRC being the problem. Hell look at the Summer reactor that was approved alongside Vogtle in Georgia: construction failure, billions wasted, and none of it to do with the NRC.
Or any the other many many other reactors abandoned at various states of development:
http://www.powermag.com/blog/nuclear-renaissance-recalls-pas...
There's an argument that the NRC could do things better, but placing all the well documented failures in the nuclear construction industry on the NRC doesn't make sense. Who are going to believe, the people who are always late and over budget, or the bystanders in the industry that have watched it all play out?
The expectation is that their close-to-Trump investors will push for the dismantling of the NRC, which is something Republicans wholly support, which will of course make their rejection moot.
DOGE told regulator to ‘rubber stamp’ nuclear[1]
[1] https://www.eenews.net/articles/doge-told-regulator-to-rubbe...
FYI (this was news to me, so it may be to other people), DOGE no longer exists. So in the end they did manage to cut some waste, themselves!
https://www.reuters.com/world/us/doge-doesnt-exist-with-eigh...
Maybe but the underlying tech still needs to perform which, as i understand it from public docs, has not. No amount of clear runway will make up for an airplane not being able to take flight.
The NRC is now mostly composed of Trump appointees. They’ve been quietly doing that. His most recent appointee was just made chair. Expect permits for friends of the Trump family and heavy regulation for competitors.
https://www.washingtonexaminer.com/policy/energy-and-environ...
Let's be honest it's one of the smartest and most useful place anyone could be investing - it's literally is whatever happens a way to contribute to mankind - even if it's just so that FB servers are off-grid ; it's still a huge win, I just hope Mark realize he has much more potential than what he is doing rn
The progress on 'nuclear' is so slow, that the same investment in Batterie and renewable would actually help a lot more around the globe.
We know how to build nuclear, we don't do it because its too expensive. Other forms are so far away from being useful, that the current Storage + Renewable pricing is so crazy good, that whatever you do with nuclear will just not be able to compete.
And the benefit? Every 3th world country and person can invest in small and big Storage + Renewable but they can't do the same with nuclear.
>We know how to build nuclear, we don't do it because its too expensive.
Refusing to build nuclear for decades makes it more expensive. If we start actually building reactors the cost will come down.
>the current Storage + Renewable pricing is so crazy good, that whatever you do with nuclear will just not be able to compete.
I would find this more persuasive if there were no new investment in carbon sources, but carbon sources have clearly remained competitive with batteries + solar, and global carbon emissions remain at an all time high. There's demand for baseload energy.
Building nuclear power stations includes a lot of labor-intensive hard to automate tasks like construction. Baumol's cost disease means it's getting even more expensive: rising general productivity leads to higher wages and higher costs in fields that cannot increase productivity as much as the general economic growth. That's why it's also still cheaper in countries with access to low-cost labor.
SMRs are a try to get out of it by building more but smaller reactors. The reality is however that nuclear has an issue with scaling down. Output goes down way faster than costs and most SMR designs have outputs far greater than what initially counted as an SMR.
Investment in renewable energy already greatly outpaces investment in fossil energy. The economic decision to keep using a fossil system is a different one than having to choose a new one. There's still problems that have no economically competitive renewable solution yet, but a lot of what you are seeing is inertia.
Base load electricity is simply an economic optimisation: demand is not flat, but the cheapest electricity source might only be able to create a relatively flat output. You'll need more flexible plants to cover everything above the base load. If you have cheap gas, base load does not make any sense economically.
For the last two years more than 90% of new power generation capacity added globally was renewable. Est 95% in 2025. So no, new carbon sources are not competitive.
https://www.wri.org/insights/state-clean-energy-charted
Highly misleading stat. That's referring to capacity expansion, not new construction.
Prior energy assets go offline and are replaced each year. The report you cite is discounting all of that, looking only at expansion above the baseline, then taking total renewable construction and calcuating renewable total construction's share of expansion. Apples to oranges.
If you look at the chart in your own link you'll see that carbon construction investment exceeds renewables still.
Chart: "Annual energy investment by selected country and region, 2015 and 2025"
I would love for what you say to be true but it just isn't, even by that agency's own stats.
Nuclear is expensive even after the reactor is build.
And I wouldn’t call it progress to still rely on steam machines for energy
What's wrong with steam?
It's better than carbon. And solar + battery requires more carbon to produce than nuclear energy as there's a lot of mining and physical construction involved + you must overbuild to supply power or rely on non solar sources.
All for building solar. Do not understand the constant need to denigrate nuclear in favour of carbon sources while doing so.
(If carbon sources were at zero this would be a different conversation)
Nothing inherently wrong with steam, just as there's nothing inherently wrong with spinning rust hard disks or punch cards.
We are at the end of the tech curve for steam, we have pushed it hard and made some super impressive technology, but it's not advancing anymore. Supercritical CO2 might have some advantages, or other fluids.
We have zero-carbon tech that uses non-steam principles, and is currently on a tech curve that's getting cheaper than any thermodynamic cycle. We have storage tech now which is an even bigger revolution for the grid than cheap solar, because a huge limitation of the grid has always been the inability to store and buffer energy.
I still have pinning rust disks, but only because they are cheap. If SSDs were cheaper, then we would see a massive switch.
(BTW denigrating steam also denigrates all fossil fuel electricity sources, because they use the same mechanism, except for some natural gas turbines)
It’s an inefficient way of producing energy. Only 30-35% results in electricity
If you believe that figure, that's still comparable to solar's best ( https://en.wikipedia.org/wiki/Solar-cell_efficiency ).
Optimal steam plants can get do better, exceeding 50% in some configurations ( https://en.wikipedia.org/wiki/Combined-cycle_power_plant#Eff... ). Steam is awesome.
What alternative do you propose that's more efficient?
Could you please provide comparable figured of EROI for solar vs Nuclear?
For a useful comparison you have to compare both sides, not give a stat in isolation and assert it is worse without comparing.
> Nuclear is expensive even after the reactor is build.
Solar panels and wind turbines need maintenance too. And they have much shorter operational lives than nuclear power plants, meaning they'll need to be expensively replaced much more frequently.
> And I wouldn’t call it progress to still rely on steam machines for energy
Could you please explain your objection to steam-based power? Is it purely aesthetic, or is there some inherent downside to steam turbines that I'm not aware of? Also, concentrated solar power systems that concentrate sunlight and use it to boil steam[1] are significantly more efficient than direct photovoltaics.
[1]: https://en.wikipedia.org/wiki/Concentrated_solar_power
> If we start actually building reactors the cost will come down.
Why would I invest then if it can't even pay for itself?
we don't do it because we forgot to do it cheap. At 3bn/unit like GE's ABWR or chinese hualongs/cap's it's a steal. At 20+bn/unit it's not that fun.
There's the "we forgot" hypothesis, but I think a more realistic hypothesis is the "we got too rich" hypothesis.
Construction productivity has stayed stagnant for more than half a century, while manufacturing productivity has sky rocketed and made us all fabulously wealthy compared to when the first nuclear reactors were built half a century ago.
I don't trust China's public cost numbers as much as I trust their actual capital allocation on the grid. And I will trust GE's numbers once they have actually produced something at those numbers, as pre-build cost estimates for nuclear are not believable due to their extensive track record.
We didn't forget how to do it cheap. The ALARA (As Expensive As Reasonably Achievable) policy simply made it illegal to do it cheaply.
Since you say ALARA made things expensive, maybe you can tell me how you foresee cheaper designs without it?
There's a lot of talk and some very shady science about getting rid of ALARA but nobody says what will change on the build that is causing the cost. Meanwhile China has adopted the same designs as in the West, without abandoning ALARA.
Those who advocate for changing ALARA see to be mostly trying to shift the Overton window on the public opinion of radiation rather than trying to pursue engineering and cost goals. I hope I am wrong on that!
Nuclear + Batteries could be nice too because the reactors will be always working at optimal rate without having to start/stop them to adapt to demand and let the storage manage peaks and lows. So investment in one domain can help the other too.
Are there targeted investment vehicles for the general public, like an ETF?
The promise of small nuclear reactors, modular reactors, thorium or whatever else has really failed to materialize at the same time that solar and battery has just leapfrogged the entire field. Nuclear has some big advantages, but it's still mired in humongous upfront costs and the intractable issue of nuclear waste. And I think we're also about to see an explosion in enhanced geothermal. The good kind of explosion.
Mark is doing much more than should already, and not the good kind, no at all!
This have a slight potential of becomeing a good one, if we only dream good things. Very limited details here, pure corporate self paise dominantly, can become anything. Another bad for example.
They're just purchasing power from existing nuke plants. All this is doing is making a formerly publicly available resource private. This will drive up energy prices for everyone else. These datacenters need to build their own generation. When people talk about how important it is that the US lead innovation in AI what they're really saying is how important it is for their quarterly results
> Our agreement with TerraPower will provide funding that supports the development of two new Natrium® units capable of generating up to 690 MW of firm power with delivery as early as 2032.
> Our partnership with Oklo helps advance the development of entirely new nuclear energy in Pike County, Ohio. This advanced nuclear technology campus — which may come online as early as 2030 — is poised to add up to 1.2 GW of clean baseload power directly into the PJM market and support our operations in the region.
It seems like they are definitely building a new plant in Ohio. I'm not sure exactly what is happening with TerraPower but it seems like an expansion rather than "purchasing power from existing nuke plants".
Perhaps I'm misreading it though.
If history repeats itself ... tax payers will be fitting the bill. Ohio has shown to be corrupt when it comes to their Nuclear infrastructure. [0] High confident that politicians are lining up behind the scenes to get their slice of the pie.
[0] https://en.wikipedia.org/wiki/Ohio_nuclear_bribery_scandal
Well, private investment is a great way to avoid subsidy nonsense.
The weasel wording is strong here. That's like me saying that buying a hamburger will help advance the science of hamburger-making. I'm just trading money for hamburgers. They're trying to put a shiny coat of paint on the ugly fact that they're buying up MWh, reducing the supply of existing power for the rest of us, and burning it to desperately try to convince investors that AGI is right around the corner so that the circular funding musical chairs doesn't stop.
We got hosed when they stole our content to make chatbots. We get hosed when they build datacenters with massive tax handouts and use our cheap power to produce nothing, and we'll get hosed when the house of cards ultimately collapses and the government bails them out. The game is rigged. At least when you go to the casino everyone acknowledges that the house always wins.
Well in a way they are building their own generation by paying elevated prices for nuclear to keep it running, as most nuclear will be shutting off pretty soon due to cheaper alternatives.
Electricity generation is getting cheaper all the time, transmission and generation are staying the same or getting more expensive. Nuclear plants get more expensive the more of them we build, but for already paid-off nuclear reactors there's a sweet spot of cheap operations and no capital costs before maintenance climbs on the very old reactors.
Meta paying for all that very expensive maintenance is not a bad deal for others, unless market structure is such that the price for entire market is set by this high marginal generation from uneconomic aged plants.
> Electricity generation is getting cheaper all the time, transmission and generation are staying the same or getting more expensive
I'm not sure what you're trying to say here, since you claim that generation is getting cheaper, staying flat, and getting more expensive all in a single sentence.
But I can tell you my energy bill hasn't gone down a single time in my entire life. In fact, it goes up every year. Getting more (clean!) supply online seems like a good idea, but then we all end up paying down that new plant's capital debt for decades anyway. Having a company such as Facebook take that hit is probably the best outcome for most.
Oops, that's a typo, should be transmission and *distrbution
Electricity costs have two components: "generation" to put power on the grid, and then the "transmission & distribution" costs which pay for the grid. You can likely see the costs split out on your bill, and the EIA tracks these costs.
Generation costs are falling, because of new technology like solar and wind and newer combined cycles natural gas turbines. However the grid itself is a bigger part of most people's bill than the generation of electricity.
Most utilities have guaranteed rates of profit on transmission and distribution costs, regulated only by PUCs. T&D tech isn't getting cheaper like solar and storage and wind are, either, so that T&D cost is likely to become and ever greater part of electricity bills, even if the PUCs are doing their job.
Generation in many places is disconnected from the grid, and when somebody makes a bad investment in a gas turbine, then the investor pays for that rather than the ratepayers. Look at Texas, for example, where even being at the center of the cheapest natural gas in a country with exceptionally cheap natural gas, solar and battery deployments hugely outpace new natural gas. That's because investors bear the risk of bad decisions rather than rate payers.
In places that let utilties gamble their ratepayers money, and where the utilities only answer to a PUC that gets effectively zero media coverage, there is a massive amount of corruption and grift and fleecing of rate payers.
A MW of nuke capacity is not replaced by a MW of solar or wind. New generation is much cheaper, but only because we are neglecting the parts of it that are hard and expensive - storage and transmission. Renewables without those things are worse than nuke - they are undispatchable like nuke and they are uncontrollably variable. We should build more renewables, but it is essential that we either tolerate intermittent system outages or massively improve transmission and storage, the generation is the least important part right now .
> New generation is much cheaper, but only because we are neglecting the parts of it that are hard and expensive - storage and transmission.
That's not correct, including storage with solar is still cheaper than nuclear. That's not measuring the cost by MW or GW, it's by measuring the cost of kWh, or the levelized avoided cost of energy, or the whatever metric you want.
And solar has the benefit of being able to avoid a good chunk of transmission by placing it at the site of use, so including transmission costs can only be to the benefit of solar.
>Unlocking Up to 6.6 GW
You could just as accurately sum it up by saying they would like to tie up nearly 6.6 GW, otherwise they wouldn't be making quite as large a deal. They wouldn't be doing it if they didn't have a financial technique to afford it, and it's still taken a while to make the commitment.
What about less-well-heeled consumers who would be better served if the effect of increased demand were not in position to put upward pressure on overall rates?
To the extent that new debt comes into the mix, that's just an additional burden that wasn't there before and this is a very sizable investment at this scale. So the compounding cost will have to be borne for longer than average if nothing else.
Naturally some can afford it easily and others not at all.
I don't really get the antagonism with these ersatz concerns. when FB builds its own datacenters, or it's own chips & racks, or it's own algorithms absolutely no one is saying "well there's no profit motive to build a completely custom server chassis" or "oh no, theyre taking publicly available math and making it private"
It's a purchase commitment, which enables the generators to secure loans to build out additional capacity.
Thank you for stating things so plainly, it's sorely needed on this site. The idea that success for big tech means a better society for workers or citizens is laughable and should soundly be rejected. They need to be broken apart yesterday.
Alternatively a more optimistic and high potential future is more plentiful and cheaper more reliable power and transmission is a huge win for society. So Id say its a perspective issue.
We in Ohio don't need more nuclear. The costs for maintaining what we have is already falling behind solar and wind (including batteries). Then there is the ecological costs that rarely get factored in.
And all these new datacenters are pushing up our electric bills. Maybe this deal could be competitive long term with newer reactor designs and if they are competently executed, but I'm very skeptical.
Maybe PA situation is different
How is more nuclear waste a huge win for society?
Well, because it means that other energy generation sources like oil, gas, and coal aren't being used there instead. Since they cause far, far more harm than nuclear waste does, it's a net win.
The same is true for solar and wind energy but without nuclear water. She cause far less harm than nuclear waste, even bigger win.
Our main problem isn’t energy production it‘s storage and quick reaction to consumption spikes. Nuclear energy doesn’t help with that.
the waste isn't a win, of course, but is a downside of a tradeoff that is massively weighted to the upsides for society -- that is (otherwise) completely clean always-on high capacity energy production.
We understand very well how to safely handle nuclear waste and make it a very (very) low risk downside.
Completely clean? Where does the nuclear fuel come from? How do we get it?
Does the handling of nuclear waste consider foul play by terrorists and such?
I didn’t he many worries about russian rockets hutting wind turbines in the war in Ukraine.
looking backwards in the supply chain for other externalities is a good point, but I'm not sure any energy production method is exempt from this?
Also, by the way, my perspective isn't about nuclear Vs X (wind turbines etc) - I like all the ones that are net clean and useful in different circumstances as part of a mix.
I'm just addressing the narrower point about whether nuclear per se is a net benefit for society, which I believe it is, massively.
I wish that Meta would pay for the extension of Diablo Canyon in California. They have had to jack up already sky-high electricity rates to keep it going, after deciding nearly a decade ago that it would be uneconomical to try to extend its lifetime.
Meta's nuclear intention is a perfect example of how tech is willing to pay far more for energy than other customers, and how it's driving up everybody's costs because we are all paying for that increase at elevated prices.
Nuclear is extremely expensive, higher than geothermal, renewables backed by storage, and natural gas. Nuclear is good for virtue signaling in some communities, but from the technological and economical perspectives, nuclear is very undesirable and unattractive. It's only social factors that keep alive the idea of new nuclear in advanced Western economies, not hard nosed analysis.
Here's a new preprint from Germans showing that even for Europe, a continent with very poor solar resources for many countries, new baseload is not the most economical route:
https://www.cell.com/cell-reports-physical-science/fulltext/...
Geothermal energy is highly location-specific (it's great if you have a handy volcano nearby, of course), storage cannot meaningfully "back" intermittent renewables because it's only good for a few hours load from the grid (aside from pumped hydro which is effectively built out), natural gas peaker plants are very expensive and increase CO2 emissions. There isn't much of an alternative to nuclear.
Enhanced geothermal using fracking techniques developed in the past decades is being deployed now, and has very limited restrictions on where it can go.
> storage cannot meaningfully "back" intermittent renewables because it's only good for a few hours load from the grid
That's trivially untrue. If you can build a battery for four hours duration, then of course you can build another to get to 8, or however many you want. Batteries are being added to the grid at a huge rate.
> natural gas peaker plants are very expensive and increase CO2 emissions. There isn't much of an alternative to nuclear.
Gas peakers are about the same cost as nuclear. We will have a ton of gas peakers hanging around in the coming decades, and they will be used less and less as we get more batteries on the grid. Already, batteries eat peakers' lunch economically.
In the most optimistic grid modeling scenarios, nuclear can play a 5% or 10% role in a fully decarbonized grid. If you go full nuclear, then you also need a ton of batteries. And if nuclear was cheap, I would advocate going for a ton of nuclear plus batteries. But nuclear is super expensive, and doesn't scale fast enough to meet our needs.
We are currently at 20% nuclear power in the US and have a rapidly aging fleet. Even if we had investors who wanted to spend the $500B it would take to keep 5% of our grid as nuclear, it's unlikely we'd be able to build those 50 1GW reactors over the course of the next 20 years. Scaling SMRs seems even less likely.
I really hope I'm wrong and the SMRs somehow materialize and are cheap, but none of the startups are acting like the have anything real or the chops to scale. A new reactor getting built by 2032, as suggested here? Pumping out an SMR in 6 years, when design isn't even finalized, the company hasn't shown progress since abandoning an NRC application, does not seem plausible.
Diablo Canyon in particular is big (https://www.pge.com/en/newsroom/currents/energy-savings/diab...). It might be too big for their balance sheet. I imagine they picked the most economical sites to expand?
Nuclear isn't extremely expensive; in China it costs around $2/watt of power (compared to up to $14/watt in the US). It's just expensive in America because America's shit at building nuclear power plants efficiently.
> Nuclear is extremely expensive
Nuclear compliance and certification is extremely expensive. The actual construction and maintenance costs are fairly trivial.
The largest cost associated with a new nuclear plant are the interest payments given that a plant may need to spend 10+ years sitting idle before it can be activated.
> The actual construction and maintenance costs are fairly trivial.
That is not at all what I have seen, the costs tend to be from absolutely massive infrastructure needed to last a long time in harsh conditions that are difficult to repair.
Those seem more like fundamental engineering requirements.
Across four different regulatory structures: France, Finland, the UK, and the US, modern nuclear has proven to be excessively expensive and require massive amounts of high skilled labor. In the past century, high skilled labor was cheaper, but these days we need to pay welders and other construction workers higher wages because they have high productivity alternative jobs that pay better than in 1970.
Those high interest payments for 10+ years are also because EPC promises to build the design within 5-7 years then takes 2-3x the time. At Vogtle the fuckups both on design meant that many plans were "unconstructable" and then construction proceeded anyway with whatever they could wing together then they had to go back and make sure that whatever the bell was built still met the design.
Construction is actually very expensive and also the plant doesn't "sit idle" for 10+ years before activation.
Construction costs are not "trivial" if each I-beam you're using has a PhD in materials testing.
What, we should have a wild west where everyone can set up their own nuclear power station without any compliance or certification? If not then these are part of the build cost... it's like saying we shouldn't include testing as part of the cost of building software.
> compliance and certification
You can say the same for cars, houses, appliances, medical devices, elevators, stairs, disabled access, etc etc.
So, what exactly is your point? Yes, everything would be "much cheaper" if nobody had to pay as much attention to most details any more. Everything would also be much much more expensive for everybody else and longer term, or not work at all or reliably or safely.
It's a question of magnitude. Do you think that over-regulation of specific technologies is possible?
If the price of building stairs was growing each year in only the west to the point were we were opening one staircase 5 every years, it might be worth to ask some companies why. If they all say "the last guy who built stairs got bogged down for 25 years trying to meet all the safety standards". It might be time to relax some of them.
If this were true china wouldnt be winding down its nuclear program in favor of solar and hydro
They're not winding down per say, nuclear has stayed steady, renewables simply exploded. Nuclear understandably more prone to schedule delays, but now it's mostly execution delays in months vs regulatory delays in years. They're still on trend with nuclear year plans. Solar simply scales much faster because all the displaced construction worker from real estate slow down can slap panels but not reactors.
They sort of are, nuke is a tiny % of their annual new capacity. Also, nukes don’t have to be cost competitive in a single party state.
That’s something that is possible only in the USA, for the best and the worst.
I don’t know any other country that would allow a truly private company to mess around with nuclear reactor.
The UK would quite happily.
Curious how this new design addresses the biggest safety challenge of nuclear reactors (the issue that was the root cause of the Fukushima accident and an indirect cause of Chernobyl): how do we ensure that the nuclear core temperature remains controlled during exceptional events (e.g., earthquakes, structural failures) when the reactor must shut down abruptly?
Chernobyl and Fukushima were different accidents and causes. Chernobyl was a systemic failure of the soviet system. Fukushima was a wild edge case that an earthquake and tsunami drained the coolant.
Edge cases don't count?
The truth is, all reactors ever built were considered safe at their time with whatever definition of safe. No one builds unsafe reactors. Yet they turned out not to be safe.
Nothing is ever perfectly safe and a lack of perfect absolute safety is not a valid objection. All sources of power have associated risks, even renewables. Wind power has 0.04 deaths per terawatt hour and solar has 0.02 [1]. Nuclear power has 0.03 deaths per terawatt hour (safer than wind), and it's worth noting that almost all of those are from Chernobyl, which was considered unsafe even at the time (they knew about the positive void coefficient). I'm not arguing that nuclear power is perfect, mainly because it isn't. But it's not like all other sources of power are idyllic havens of safety. There are always tradeoffs.
[1]: https://ourworldindata.org/safest-sources-of-energy
Those figures seem very optimistic. Uranium miners die early, often of horrific cancers.
https://pubmed.ncbi.nlm.nih.gov/33232447/
But the bottom line is that renewable costs are trending down, hard and fast, battery tech is just getting started, and development time for wind and solar is comparatively fast.
Future nuke costs at this point are speculative, development time is very slow, and even if new reactors were commissioned tomorrow, by the time they came online it's very, very likely solar and wind + storage would make them uneconomic.
IMO the attachment to nukes is completely irrational. There are obvious economic downsides, no obvious economic benefits - and that's just the money side.
Thanks for the reply! I think you're arguing with the wrong person in the second half, though. I agree that renewables could potentially be more economically viable than nuclear power[1]. My reply was disputing the "people can die from nuclear therefore we should never use nuclear" argument, not arguing about economic viability. Also I think that broadly claiming that your opposition is "completely irrational" is not a very tactical rhetorical move.
[1]: although since you're basing your claims on the speculative future state of solar technology 10 years in the future, I don't see why the same shouldn't apply to the speculative future state of nuclear power, but that's besides the point
What mechanism causes solar power deaths?
Apart from the deaths from workers falling off the roof or from wind turbine towers (though these might be the only type of deaths included in these figures):
If mining deaths are included, coal, oil, gas and uranium probably do not look favorable at all, but renewables aren't perfectly safe either: there was a bridge collapse at a copper/cobalt mine in Congo two months ago that killed 32. Solar and wind use more copper per energy unit than other technologies, and solar and wind indirectly require battery technology. Lithium batteries contain lithium and cobalt. (Lithium mining seems relative safe, but 70% of cobalt is mined in Congo, which is known for artisanal mining, and the above-mentioned accident indeed seemed to happen at such a mine.) Wind, especially off-shore wind uses more concrete and steel than other power generation technologies (hydro seems like it'd use a lot too?), which could be explored too. (Course, these metals are recyclable, so you only mine them once.)
Battery factories also produce deaths sometimes, e.g. recently https://en.wikipedia.org/wiki/Hwaseong_battery_factory_fire, and batteries in operation as well as discarded batteries sometimes produce deaths too.
Accidents, mainly. Solar panels and wind turbines produce far less energy per module than nuclear, so you have to build much more of them. If you build enough of something, the odds that everything goes perfect every single time are quite low.
They count as a different type of failure.
We knew what to do but screwed up hard is a operational failure and we didn't plan for it is a design/planning failure.
The people who are hurt might not care, but understanding the root cause is important to address them.
> Edge cases don't count?
Constrained edge cases are fine. Particularly when contrasted with coal and natural gas, which are, in practice, what everyone is competing against in America.
The edge case was predicted, but market and political forces chose to ignore it. The GE Type II reactor had known issues. [0]
I should add that I am not strictly anti-nuclear, and it is super interesting that some of the largest funders of anti-nuclear propaganda have been actors from the fossil fuel industry. [1]
[0] https://publicintegrity.org/environment/reactors-at-heart-of...
[1] https://www.forbes.com/sites/kensilverstein/2016/07/13/are-f...
Wild edge cases are to be expected when you do things at scale. If you build 20 buildings in different regions, at least one of them will likely face a once-in-1000-years natural disaster. And it's difficult to estimate how bad that particular kind of once-in-1000-years event can be, because you probably only have a century or two of reliable data.
Let’s not forget that you don’t have to be a socialist/communist nation to decide you want to do the cheaper thing. Without robust regulations I guarantee you a Chernobyl-like disaster could easily happen in the US because of less scrupulous companies cutting corners and choosing the cheaper path. With Chernobyl it was the government instead of a private company.
We can talk all day about how the system incentivized people playing CYA rather than actually trying to solve the problem (true and fair critiques), but when it comes down to it, this happened because the cheaper option was chosen and potential issues were overlooked. That transcends political systems.
I personally find highly hypothetical situations impossible to guarantee but I'm glad you have such a high degree of self certainty for a plausible scenario you have decided to give certain results to.
You should really consider educating yourself on the Chernobyl reactor melt down (read a book or two) to understand the level of calamity inflicted by the communist system. Stop trying to make it sound like that could happen anywhere because the pressures of capitalism could cause the same results. Its pretty eye opening how insane the chernobyl situation was.
The US had Three Mile Island. Japan had Fukushima.
One of the biggest arguments against nuclear is that reactors are insanely complex. Beyond a certain level of complexity, safety and predictability become impossible even with perfect management - which certainly doesn't exist in the nuclear industry.
This is especially true of any nuke system which needs external cooling, because stable water levels aren't a given any more because of climate change. Between floods, droughts, and storm surges, the environment is part of the system - something Fukushima discovered to its cost.
I am actually very familiar with the history of Chernobyl and the meltdown. What I’m saying is human greed and short sightedness do not suddenly go away because a nation decides on a different political/economic system. The implication that it only happened because it was the Soviet Union is what I’m taking issue with because it absolutely could happen in the US without proper guardrails. All it takes is one bad company cutting the wrong corner or firing the one person who spoke out. It’s very easy to see no society is immune to this.
I am not defending the Soviet Union or any of the decisions made during Chernobyl. So you should redirect your indignation/condescension.
You don't need to prevent it. You just need to prevent a catastrophe and even Fukushima did it relatively well - nobody died or will die from radiation. Current benchmark for (future) gen4 designs is having consequences limited to the area of the plant, think of 3MI but as worst case. But imo it's still an overkill, nuclear is one of the safest sources in terms of human deaths/kwh and the stat only gets better with gen3/3+
> You just need to prevent a catastrophe and even Fukushima did it relatively well - nobody died or will die from radiation.
“As of 2020, the total number of cancer and leukemia instances has risen to six cases according to the Tokyo Electric Power Company (TEPCO).[5] In 2018 one worker died from lung cancer as a result from radiation exposure.”
https://en.wikipedia.org/wiki/Fukushima_nuclear_accident_cas...
These are small numbers compared to the number that died due to the tsunami and the massive evacuation (to avoid radiation injuries). The frustrating bit is that they could have avoided it all.
The linked to article makes a different claim
" The workers' compensation claims that have been recognized by labor authorities include six cases of workers who developed cancer or leukemia due to radiation exposure "
So compensation has been requested for cancers, of which one death has been reported.
I point to a Forbes opinion piece from a pro-nuclear person https://www.forbes.com/sites/jamesconca/2018/09/06/no-the-ca... .
I remember when terrapower was supposed to build a kind of reactor which would use depleted/unenriched uranium which was basically a hole in the ground slowly burning up, but now it seems they're doing a molten sodium "classic" reactor, is that the same thing and I am confused, or did they change their strategy?
For reference, China installs about 1 GW of solar per day. By this time next week, they will have surpassed the output of this entire project.
China is the world's largest electricity producer and installs a lot of generating capacity of all types. For example, China has 29 nuclear reactors with 31 GW of capacity currently under construction:
https://pris.iaea.org/PRIS/CountryStatistics/CountryDetails....
Which leads to a shrinking nuclear share in their grid. It peaked at 4.6% in 2021, now down to 4.3%.
Compared to their renewable buildout the nuclear scheme is a token gesture to keep a nuclear industry alive if it would somehow end up delivering cheap electricity. And of course to enable their military ambitions.
I think that says more about their vast investment in other forms of power (particularly renewables) than it suggests a lack of investment in nuclear.
The nuclear share dropping is a very clear signal about a lack of investment. Shows that nuclear energy is no longer cost competitive, even in a "low regulation" environment.
It shows that strategic investment matters and people are looking at more than a single cost metric. Nuclear is behind today, but that doesn't hold a promise it will remain true into the future unless you stop investing now.
One armed bandit says explore as well as exploit. This delta you cited indicates the pendulum currently is more exploit than explore, but its not a static equation.
chinese nuclear is extremely cost competitive at 2.5bn/unit. They have other reasons, one being the ban on inland expansion fearing of messing up with 2 major rivers that feed the country. Current chinese units are basically borrowed and improved western designs, cap is basically vogtle's ap1000, hualong is a frankenstein of several western designs.
>borrowed and improved western designs
TBH this part seems key, even PRC couldn't operate full western designs reliant on western industrial capacity economically, part of it was simple incompetence of western supply chains (business closures / regulatory drama / sanctions). Nuclear seems viable once you strip out a lot of the politics that makes them uneconomical, hence PRC had to indigenize the designs since once western supply chains enter picture, the schedule goes out the window.
The US, in 2024, installed ~0.13 GW of solar per day.
https://seia.org/research-resources/solar-market-insight-rep...
6 GW Nuclear is either a tech company getting ahead of bad PR with a token gesture. Or its maybe? the start of something real.
Two years ago we were installing 1/10th of Chinese solar today?
Where are we at today? Can we catch up under this administration?
Where do we compare on a nuclear basis? I know my state installed nuclear reactors recently, but I'm not aware of any other build outs.
In a war game scenario, China is probably more concerned about losing access to oil and natural gas than we are. Not that we shouldn't be building this stuff quickly either.
> Can we catch up under this administration?
No. The future is Chinese, if the Chinese can maintain good governance.
A big "if"
1 GW in nameplate capacity of solar panels powers fewer data centres than 1 GW of nuclear. So it needs to be "this time next month".
1GW at noon, maybe 20% of that average.
China's building a bunch of nuclear too.
That isn't how solar capacity is measured. It's not simply its maximum instantaneous power potential
There are multiple measures, as generating technologies are complex. "Nameplate capacity" (given above) is one, "capacity factor", which is (roughly) the time-averaged output is another, and for solar averages about 20%, though that can vary greatly by facility and location.
Nuclear has one of the highest capacity factors (90% or greater), whilst natural gas turbines amongst the lowest (<10% per the link below). This relates not only to the reliability of the technologies, but how they are employed. Nuclear power plants cannot be easily ramped up or down in output, and are best operated at continuous ("base load") output, whilst gas-turbine "peaking stations" can be spun up on a few minutes' notice to provide as-needed power. Wind and solar are dependent on available generating capability, though this tends to be fairly predictable over large areas and longer time periods. Storage capability and/or dispatchable load make managing these sources more viable, however.
<https://en.wikipedia.org/wiki/Nameplate_capacity>
<https://en.wikipedia.org/wiki/Capacity_factor>
It's close enough to how it's measured. China's terawatt of solar power capacity isn't producing 9000 terawatt hours in a year. Their total electricity use is 9000 terawatt hours.
It is how individual power generation projects and measured though. If you install a GW of solar generation, it means you installed solar panels capable of generating 1 GW peak. If you install a 1 GW of coal generation, then same thing. If you install 1 GW peaker gas plants etc.
The coal plant will have a capacity factor of 80% though. Solar will be 10 to 20%. And the gas plant could be very low due to usage intent.
Battery projects are the same (since they're reported as generators). Whatever nameplate capacity...for about 4 hours only.
As mentioned numerous times in this thread, the percentage of their generation that is nuclear is falling.
And yet they're still making a bunch more.
So only installing 73GW average capacity per year.
Yes. Meta's matching a whole month of China's solar growth, which I would call a lot.
Solar and Nuclear energy are different energy products. China is also bringing on an insane amount of nuclear energy.
In absolute terms, China installs about as much nuclear as the US does solar. So I can only assume you agree with the statement "the US is bringing on an insane amount of solar energy"? Because, once again in absolute terms, the US's solar buildout is trounced by China's. The US is losing the energy race, and nuclear isn't going to save it. The US will run out of fissile material before China runs out of sunlight.
Depends if they start seeding clouds and doing geo engineering.
Nuclear and solar are different energy products that are complementary. This solar vs nuclear narrative is basic and anti progress.
For example china invested in solar so they can transition their energy system and get it paid by selling globally via subsidized cell manufacturing.
I don't think they will be able to sell export their nuclear tech globally since it is essentially repackaged US tech.
But yeah Im all for solar - more solar the better but it cant do firm power well.
There's really no risk of running out of fissile material. We can create it.
China is building a tiny amount of nuclear in comparison to their wind, solar, storage, and HVDC builds. Only something like 50-100GW over thw coming decades. The quantity being built only makes sense as a strategic hedge, not as a primary strategy.
Renewables crash the money making potential of nuclear power. Why should someone buy ~18-24 cents/kWh new built nuclear power excluding backup, transmission costs, taxes, final waste deposit etc. when cheap renewables deliver?
https://oilprice.com/Alternative-Energy/Renewable-Energy/Wha...
China is barely building nuclear power, in terms of their grid size. It peaked at 4.6% in 2021, now down to 4.3%.
Compared to their renewable buildout the nuclear scheme is a token gesture to keep a nuclear industry alive if it would somehow end up delivering cheap electricity.
nuclear provides for about 4-5ct/kwh if built cheap, everything included, looking at swiss data. Chinese units are built for 2.5bn/unit, so probably even cheaper than that. But yes, china is far from what france or sweden did with nuclear per capita
Again they aren't the same product. Everyone always thinks power is only about $/kwh especially in hackernews. That is a strong proponent of the product but most definitely not all of it. Solar just does not work for large scale industrial uses cases (99.99% uptime). Even with massive energy storage to try and cover the edges. Its a great combo but not comparable.
How does your "large scale industrial use case" deal with 50% of the French nuclear fleet being offline?
https://www.nytimes.com/2022/11/15/business/nuclear-power-fr...
Or 50% of the Swedish fleet two times this year being offline?
At the same time it happened, the french solar was 92% offline and the french wind generation was 81% offline.
Maybe we should get the opposite conclusion from this incident.
Based on yearly average capacity factor?
Since that incident storage has been scaling massively. How does a nuclear plant compete with zero marginal cost renewables?
https://oilprice.com/Energy/Energy-General/The-Quiet-Unravel...
by providing firm power.
Based on if it would run at full capacity like a plant would yes.
And no, storage hasn't scale at all yet, it would need a 100x increase before being useful for such events.
The proof is in the pudding anyways, if that works so well, why nobody does it?
Nobody does what? Solar installs are way way up.
Nobody does a storage based solar grid at a country scale.
Renewables and battery storage energy are unstoppable. Why take nuclear risk when you can get more than enough from solar, wind, and geothermal coupled with battery storage?
lol, talking about risky nuclear investment and mentioning geothermal as ready to go alternative...
It’s not easy to go. It’s going, and has been for many decades.
Small scale where I am, when compared with hydro.
Geothermal is essentially negligible to the grid so it's weird to include it.
It's also so geographically constrained no one can choose to build it anyway.
China is a country with over a billion people, Meta is a private company with under 100k employees, it doesn't really make sense to compare the power output of their investments.
Nuclear power from the company that moves fast and breaks things?
Meta (seemingly) bought the local race track in Beaver PA for a large data center. We were sad to see a world class racing facility go away as it finally started to hit its stride. But the offer was too hard for ownership to pass up
Racetrack memory is an emerging alternative that might obviate the current DRAM shortage especially for datacenter use, so in a way this makes a whole lot of sense.
https://www.ibm.com/history/racetrack-memory
Is that word salad?
It's called a "pun", albeit maybe not a very good one.
Isn’t that just quantum bubble memory?
Good bot.
I'm kind of surprised there's so many nuclear power companies. I remember Google signed a deal a year or so ago [0] with a different one (Kairos). I wonder why they're not in this deal? Is it all connections and funny money? Or do any of these have real shots at making something useful?
I was thinking how the dotcom bust left a lot of dark fiber infrastructure which helped the internet take off after that. It would be great if the upcoming AI bust (if it happens) leaves a bunch of power generation and new nuclear tech behind.
[0] https://blog.google/company-news/outreach-and-initiatives/su...
The dark fiber analogy doesn't quite work though, because unused fiber just sits there harmlessly. Nuclear plants have massive decommissioning costs ($280M-600M+ per plant), ongoing waste storage requirements, and if the companies go bust, taxpayers are likely on the hook. The EU alone is underfunded by €118 billion on decommissioning. Meanwhile batteries and solar keep getting cheaper every year without the “who pays to clean this up in 50 years” problem. Seems like a very roundabout way to hope for public benefit.
with small companies it's like with startups - some might fail so maybe it's better to have more players. I still think this is strange - if they wanted nuclear - could have approached GE for some BWRX of ABWR or W-house & KHNP for some nice AP1000/alike units
Right. Meta wants big enough plants that an AP1000 or two would be the right size. They're known to work. There are four in operation, two in the US, and another dozen or so under construction.
Most of the small nuclear reactor startups hand-wave the failure modes and argue that they don't need the hulking big expensive containment building. NuScale claimed that. They wanted multiple reactors sharing the same cooling pool. If they ever had a leak, the whole set of reactors would be contaminated, even without a meltdown.
If we look at the big reactor accidents so far, there's Chernobyl, with no containment building. There's Fukushima, with too small a containment unable to contain the pressure. And there's Three Mile Island, where a large, strong containment building contained a meltdown. Three Mile Island was an expensive disaster, but not hazardous outside the plant. That's the failure mode you want.
We might be better off at developing better techniques for welding thick sections to make hulking big, strong containment vessels. There's been progress with robotic welding of thick sections.[1]
[1] https://www.agrrobotics.com/trends-s-industry-analysis/roadm...
> surprised there's so many nuclear power companies
This is actually American capitalism working at its finest.
Have you seen a video of a slime mold "solving" a maze? It reaches out in every direction with thin tendrils until it makes contact. (Then the game shifts.)
We have a sense, like a slime mold picking up on the "scent" of food, that there is energy. But there are lots of good hypotheses for how we get there. So we try them. Not exhaustively. But multiply. When someone demonstrates they've got it, the game will shift to consolidation and scaling.
This may just be me... but is anyone else in their head saying: "whats the catch?"
~~Meta~~ Facebook has made their money by de-stabilizing people's emotional state to keep them engaged and buying stuff via ads. I'm having a tough time connecting the dots between that and nuclear power.
"AI" is in the title. It's to power their data centers.
I get that and scanned the article. they are currently, and have been, working on "AI" for a bit, but strategically they aren't positioned to provide a service offering right now. Even if they are going all in and they got regulatory love, it would make way more sense to go after an energy source/provider that could provide it on a faster timeframe. they'll get their energy in what? 7-10 years?
In terms of $ per unit of energy, would shutting down Meta give best bang for buck?
Power prices would presumably fall.
So they can power all the AI data centers.
I won’t take Meta’s new ambitions serious unless Zuckerberg does something like have a “fireside chat” next to Ezra Klein moderated by Kara Swisher where he awkwardly spams the word “abundance” while staring Klein square in the brow. And then buys Vox Media.
All that energy will surely translate into quality of life for the rest of us.
It likely will.
Not immediately but in a decade or two.
Infrastructure investment is a good thing, and almost always strictly better than not having it.
"Infrastructure investment" is not fungible. There's either demand for this power where they're adding it, or there's not. If there is demand, then it's created by AI. If there's not (i.e. AI bubble pops), then there's excess capacity where we don't need it.
Can you or anyone else please take us through this likely outcome?
I think the arguments for this would possibly be, if AI continues to be useful (generation demand skyrockets): Meta would possibly have a positive ROI for these investments which would lead to others copying the investment strategy and building more nuclear. If that happens a large portion of AI demand would become green(-ish) energy.
If AI demand lowers (generation demand plummets): Meta would have subsidized a bunch of nuclear reactors which would likely continue to produce power for 10 years - 50 years.
A big reason I have heard for lack of nuclear build out is the lack of starting capital but after they are built they are generally stable and maintenance is predictable.
An example of this: https://en.wikipedia.org/wiki/Beznau_Nuclear_Power_Plant. It may be turned down eventually but a 60 year runtime is pretty impressive for 60s engineering!
> A big reason I have heard for lack of nuclear build out is the lack of starting capital but after they are built they are generally stable and maintenance is predictable.
I have also heard this, but given Meta's announcement is mostly in funding and extending the useful lifespan, doesn't that indicate without an infusion of capital, the ongoing operations are not cost effective?
I'm glad that individuals with no democratic control are allowed to declare large swaths of society to suffering because, maybe, in the future things might get better.
FWIW, it took nearly 150 years for commoners to benefit from the industrial revolution. The idea that I must suffer and my children must suffer and their children must suffer so some future plutocrat can get a fatter nut is pathetic.
I mean, we could just give up and go back into the swamps that we crawled out off millions of years ago. But I'd rather have more clean power come online. Energy is EVERYTHING.
It's not a good thing when said infrastructure produces ridiculously dangerous waste & fallout risk though.
"Fallout risk" and "dangerous waste risk" are such non-issues that have been so absurdly elevated in the public perception that the fear of them has caused far more deaths than the risks themselves.
How exactly? Meta isn’t doing this out of generosity to society. They’ll be consuming this and vastly more energy to ultimately increase their own profits?
They were being sarcastic.
This is cool, but I'm waiting for the announcements for datacenters to start establishing on-site geothermal power for baseload power supply.
https://www.businesswire.com/news/home/20250612778008/en/XGS...
seems like they're getting the ball rolling, will be interesting to see how this scales
To do what Grok has been doing?
Yikes.
I'm glad somebody is doing it, even if it's Meta. The world really needs more energy and nuclear is a great option
Is it though? Do we need more ads and more social media company AI?
As technology improves, we have less and less need for nuclear. The continent with the greatest need for nuclear is Europe, and these German grid modelers have taken a look at the EU grid with the latest data and decided that additional baseload generation (like nuclear) is not required and will likely increase costs if built:
https://www.cell.com/cell-reports-physical-science/fulltext/...
Germany took it's last three nuclear reactors offline in 2023 and now the primary source of their electrical generation is coal.
See https://world-nuclear.org/information-library/country-profil...
That is factually incorrect. The primary source is wind at 132 TWh in 2025, followed by solar with 70 TWh.
Lignite was third with 67 TWh and hard coal sits at 27 TWh.
https://www.energy-charts.info/downloads/electricity_generat...
Lignite is coal, so that'd make coal #2
Official source for 2025 Q3: 64,1% renewable 20,6% coal 12% gas
https://www.destatis.de/DE/Themen/Branchen-Unternehmen/Energ...
In addition to the other corrections here, I'd like to add one more remarkable fact: in 2025 the share of German electricity generated by solar increased to 18% from 14%. That's in a single year, in a country with terribly low levels of sun! Nuclear generated 5% of electricity before it was shut down, and had generated that same percentage for more than a decade (that's as far back as the chart I saw went).
It's remarkably easy to scale solar to very large amounts in short time periods. Far easier than building a new nuclear fleet.
Your claim about current electrical generation is incorrect and obviously not supported by your source, which shows data from 2021.
hell yeah!
While I'm sure that the vast majority of this energy is going straight to AI server farms, I'm also hopeful that this will renew efforts to switch to sustainable energy sources.
The entire world is switching to sustainable energy at a tremendous rate because they are the cheapest source of electricity, and storage is economical on every grid's peak price to low price swing that I have seen. In places like Pakistan solar is taking over because the grid made really poor investments in very expensive fossil fuels. In Africa, a panel and a battery is cheaper to run than a generator, which is life changing. As that scales up it will revolutionize entire economies.
In the US, the vast majority of new generation is renewables, matched with a ton of storage. There's some gas too, but it will be uneconomical to run these gas plants before their end of life:
https://www.pv-magazine.com/2025/08/21/eia-projects-record-6...
The biggest impediment to cheap sustainable electricity is political and basic ignorance of the voting public that allows utilities to continue installing familiar, but more expensive fossil generation.
what has home solar to do with country's grid and industrial needs? No country has a VRE+BESS solution to fully sustain itself. South Australia might get close having extremely nice weather patterns but even they will periodically import from fossils neighbors. Nuclear was dirt cheap in US but it didn't help to expand it. and now it's not that cheap since supply chain is gone
> what has home solar to do with country's grid and industrial needs?
Why are you asking this question without answering it? Unless you're assuming I said something I don't, I don't understand the relevance here.
> No country has a VRE+BESS solution to fully sustain itself.
A couple of fallacies here: non existence yet doesn't mean it's not possible, especially for a grid with equipment lifetimes measured in multiple decades. Once it makes sense to build a VRE only grid it will take several decades for it to happen.
> Nuclear was dirt cheap in US but it didn't help to expand it. and now it's not that cheap since supply chain is gone
Nuclear was completely overbuilt, based on bad over predictions for energy demand, leading to excess orders, then the only plants still being under construction being the ones that were poorly managed, leaving a distinctly bad financial taste in the utilities' mouths. They had to be bribed by state legislatures in South Carolina and Georgia to let them pre-bill ratepayers and also put ratepayers on the hook for overages on construction for new reactors, which everyone expected and which happened disastrously.
It's really unclear that nuclear is ever going to be cheap when its main costs are high skilled construction labor, concert and steel. None of those inputs are getting cheaper and the only chance at making logistics and management better seem to be small modular reactors, which are inherently cost inefficient due to losing the scale advantages of 1GW reactors.
It's no longer 1970, and the costs and technology of the modern economy are fundamentally doffeeent. We have learned so much, and I hope that we have learned tha nuclear doesn't fit with our advanced technology of the 21st century.
I'm sure some people were hopeful that the proliferation of fast food would lead to more exercise and healthy eating to balance it out.
Monopolies are insane. Why is myspace working on nuclear power?
I despise Facebook and all that it stands for, but if the surplus value that it has extracted from humanity over the last two decades is reinvested intelligently into nuclear energy, I'm actually okay with it.
Despite the hype that you see on Twitter, the hard tech startup scene is actually incapable of large-scale engineering coordination on the level needed for a nuclear power plant, or even a gas turbine.
If any innovation on fission reactors is going to be successfully commercialized, we will need to see billions of dollars of investment over medium to long time horizons.
Of course, the millstone around the neck of nuclear power is that it's a dual-use technology. There's probably a lot more behind the scenes that's been done to stifle the industry effectively for non-proliferation reasons, but masquerading as cost, regulatory problems, environmental concerns, etc.
The biggest barrier to nuclear energy in our country is the linear no-threshold safety model. We know the model is wrong, opens up operators to infinite liability, and since adopting it the nuclear industry has effectively been dead.
Even just a basic common sense threshold would make nuclear more viable overnight.
it's one of but not biggest. Biggest is poor supply chain. Try copy rosatom first in their variety of services...
Agreed, but I don't think that dumping ALARA will fix nuclear cost. There's still plenty of stuff in a nuclear reactor hot enough that we have to worry about keeping it from being inadvertently released.
So, the result will be rising energy bills for everyone?
So is the metaverse completely dead and over with?
Extremely awesome. If you can't praise this in spite of however you feel about AI, or about Meta, you need to reevaluate your priors.
Some Shreck Power vibes.
Do any of their vendors have existing products or is this a Boom Aerospace situation?
Yes. Vistra is the largest power generator in the US [1]. Most of the deal that Meta announced is with Vistra, and the power will come from some existing facilities.
[1] https://en.wikipedia.org/wiki/Vistra_Corp.
I am pro-nuclear power, but I miss the days when companies would, you know, return their profits to investors, so those investors could then invest in other companies doing different things, instead of all corporations tending towards generic everything-investment vehicles.
This is partly due to tax policy. Investors don't want this. Distributing dividends is less efficient capital distribution because those capital gains are immediately taxed rather than be deferred into a higher share price with no tax drag.
Oh no, capital would be taxed!
Don't threaten me with a good time.
This is why every tech company does buybacks(other than meta which started a dividend when Mark got sick of high interest rates on his debt, they still mainly do buybacks though)
This is to appease the current US administration running the federal government that for some odd reasons is abhorrent to the idea of using cheap, cleaner renewable energy and battery storage to decarbonize our air and grid.
After reading through all press releases I still haven't seen any details on any money changing hands?
The previous ones from Google and Amazon at least specified that it was based on PPAs. Where all they did was bind themselves to buy X amount of power at Y cost if the company could deliver.
Taking a step back I have an incredibly hard time seeing how new built nuclear power will cope with renewables fundamentally reshaping our grids.
We're seeing the unraveling of the grid monopoly infront of our eyes [1] and renewables are set to completely crash all "baseload" markets. [2] Likely forcing them to become stranded assets.
[1]: https://oilprice.com/Energy/Energy-General/The-Quiet-Unravel...
[2]: https://oilprice.com/Alternative-Energy/Renewable-Energy/Wha...
So now we’re harnessing a the dicy and capital intensive power of the atom to create the illusion of speaking to people because we can no longer strike up a conversation with a stranger on the street. Cool, cool. Personally I’m going to stick to real people with no divergent neutron chain-reactions involved.
If they want this to work trump and co gotta be out asap
Theyll just steal it
This is a purely political move. It will take a decade or longer if ever to ever get power from this. And yes it says "as early as 2032" but we know how that goes.
Why nuclear? Because it's cleaner than fossil fuelds but appeases the administration because it isn't wind or solar, which would immediately solve any power generation problems.
You might be tempted to say, since this always comes up, "what about base load?"
FFirst, batteries can solve that problem.
Second, you use a mix of power and when the Sun isn't out (ie night) is when power is cheaper from other sources.
Third, data centers don't really need base power at all. You just run the DC when you have power and don't when you don't. There's precedent for this. Google has a DC in Scandanavia that they shut down a few days a year when it gets too hot, otherwise it's just cooled by ocean water.
What I find most funny about all this is that all these big tech companies are kowtowing to the state in the exact same way they accuse Chinese companies of doing.
> Google has a DC in Scandanavia that they shut down a few days a year when it gets too hot, otherwise it's just cooled by ocean water.
They do? Which facility is this? I'm quite surprised to hear this would happen, in Scandinavia of all places.
wind and solar need firming. Currently firming capacity in US is getting scarce. It can be fixed a bit with better transmission but otherwise you need to expand it. China is expanding firm power with coal, gas and nuclear. Germany will expand gas.
Bess will not solve the firming problem. And no, if you build a multibillion datacenter you want to run it around the clock as much as possible. But yes, some datacenters don't have such requirements, but here we are talking about meta
if you think it's near impossible to reach a real human for a difficult customer service problem now, imagine by the end of this decade
I also think nuclear energy is responsible for customer service.
usually don't have to explain steps in thinking on HN
what do you think the end result is for the amount of machine-learning in use once the resource barriers go away?
like first the power availability limit, next cpu+memory availability?
result: every human job goes away that can be turned into a survey of questions
like all callcenters
sure, let's build more energy sources with finite fuel supply and negative environmental impact while there's better options available <.<
These companies abandon software? Does anyone really think this is going to end well?
Are there regulations in the US that you need to fund decommissioning up-front? I think that's the case in the UK at least.
Yeah, there's a lot of stuff for nuclear. I wrote to the government to get them to remove 10 CFR Part 20 Subpart E financial requirements but obviously that hasn't happened. You can start from there and branch out.
Don't think so but after seeing electricity prices rise and even double in areas by giant data centers many areas are requiring them to provide their own power
Nuclear energy company valuations are in the puny low ~100s of billions.
But Meta can be sued for so much more ...
Remember prior to the AI boomb electrical bills were still high.
Now we see more demand and the building of the grid to meet that demand....yet the high cost is still remaining the same.
Most of the cost of residential electricity is the distribution network, not generation. Supply/demand dynamics only really come into play during extreme scenarios like the Texas winter outage a few years ago.
How much increase in consumption is due to AI ?
The price isn’t the same, it’s more expensive, and it’s going to get worse before it gets better. This particular project is slated to take a decade — the last nuclear reactor that the US built was also slated to take a decade. It just opened in 2024, and was approved for construction in 2009. This is a long-view solution to a right-now problem.
We should have been keeping up with this infrastructure stuff all along… but I’m really not convinced all of these companies are going to be using this shit in 5 years, anyway.
I've begun to realize that the prices of everything seem artificially set to lock us into some form of farming. Carefully balanced.
Look for home prices, doesn't matter where in the US it's always like $350-$400k. Why?
Why do you think home prices are always in that price range? That's not been my observation.
Stares in Seattle home market.