Meta announces nuclear energy projects
(about.fb.com)217 points by ChrisArchitect 5 hours ago
217 points by ChrisArchitect 5 hours ago
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.
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...
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...
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.
>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.
China are building dozens simultaneously, and even with their questionable workers rights, safety and environmental practices, they cost $7 Billion a pop.
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?
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.
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!
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.
Are there targeted investment vehicles for the general public, like an ETF?
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.
>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"
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
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.
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.
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.
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.
Meta's nuclear intention is a perfect example of how tech is willing to pay far more for energy than other customers
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.
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.
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.
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.
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... .
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.
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.
1GW at noon, maybe 20% of that average.
China's building a bunch of nuclear too.
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.
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.
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.
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.
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.
It's called a "pun", albeit maybe not a very good one.
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.
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?
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.
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.
It's not a good thing when said infrastructure produces ridiculously dangerous waste & fallout risk though.
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?
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
I'm glad somebody is doing it, even if it's Meta. The world really needs more energy and nuclear is a great option
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...
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.
A core assumption of capitalism is that when individuals act in their own self-interest, their actions tend to produce outcomes that are beneficial for society as a whole. This seems like a compelling piece of evidence!
Well, no, I think that the claim is that having nuclear power plants is better than not having them. If they're not sucking energy off the grid (like what is happening right now), that at least will help avoid regular people like us having to pay the increased prices and indirectly subsidizing them.
And nuclear energy is clean (from a climate change perspective at least), and so if they're going to keep spending huge amounts of energy AI training anyway, it's probably better to do that in a way that isn't going to keep boiling the planet.
Also, if there is any kind of excess energy then it can be fed back into the grid, meaning that grid power can be fed from something relatively clean compared to something dirty (like coal).
I'm not entirely sure how this relates to the party thing. I'm saying that sometimes something selfish in a capitalistic system can occasionally still be a net good. I didn't think that was controversial. I'm not saying we give Zuckerberg a trophy or anything.
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.
Monopolies are insane. Why is myspace working on nuclear power?
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.
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.
So, the result will be rising energy bills for everyone?
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.
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.
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)
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.
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
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
These companies abandon software? Does anyone really think this is going to end well?
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.
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?
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?