HN.zip

87% of the Anthropic power will come from gas and 11% from coal in 2028

717 points by - 752 comments
[3 hidden]5 mins ago
Anthropic is a legitimately bad actor here, particularly in leasing Colossus 1 given the local pollution it has caused in Memphis. It is particularly shocking that they've made no renewable commitments in the US - Alphabet has made so many, and has a huge team on this! Just clearly not a priority, and they should be shamed and pressured. They know better.
[3 hidden]5 mins ago
How much energy are we talking about here? 300MW...? An absolute drop in the bucket compared to pretty much anything. Colossus 1 consumes about 7,165 gasoline cars continuously driving at 60 MPH. Since elon has deployed hundreds of thousands of electric cars, i would say he's net even. Especially if you count the amount of battery power and solar they've deployed worldwide. Lastly, they are actively working to solve the energy crisis by putting datacenters in space.
[3 hidden]5 mins ago
As Kant puts it: "Act only according to that maxim whereby you can at the same time will that it should become a universal law."

I don't want to live in a world where everyone computes their own energy need, decide it is an "absolute drop in the bucket" and chooses the absolute worst option to satisfy said need.

[3 hidden]5 mins ago
> Alphabet has made so many

Didn't they walk away from their renewable goals the moment that AI took off and then deleted previous pledges from the corporate website?

[3 hidden]5 mins ago
Unfortunately, as long-termists, they are happy to do short-term harm in pursuit of what they believe is a long-term good. It would be nice if they were regulated back onto a more sensible path, but not holding my breath.
[3 hidden]5 mins ago
It would be even nicer if we directed our attention to the actual problem: power monopolies posting steadily increasing profits for the last decades.

This isn't just an anthropic problem, and they're an insignificant, just stupid to even talk about it, rounding error to our energy usage. Especially when you include ICE engines as energy sources, and the many multiples grid capacity required to convert everything to electric.

This is a pebble in the ocean that has a strange and disproportionate, probably lobby based, amount of attention directed to it, in the news cycle and social media.

[3 hidden]5 mins ago
yes that's exactly what they're doing here, and that's why serious regulation is needed to force them to act like normies. the longtermism is delusional
[3 hidden]5 mins ago
that just sounds like giving someone a pass, and at this point willingly because I haven't heard anything from dario about making the world a better place other than fear-mongering
[3 hidden]5 mins ago
That's what the parent comment says. Anthropic believes they have the pass. And that's not a good thing and also that's why having laws matters.
[3 hidden]5 mins ago
They’re not called Misanthropic for nothing!
[3 hidden]5 mins ago
imagine the reactions of the people working at these companies if someone plopped 50 gas turbines in their backyard.
[3 hidden]5 mins ago
This is why we need regulation. Without rules, the cost of the externalities is passed on entirely to communities very far from leadership decision makers or shareholders.

The problem with a purely financial accounting is it ignores environmental and social costs entirely.

[3 hidden]5 mins ago
> This is why we need regulation.

The Senate passed a bipartisan bill to speed up the approval process for data centers and power plants. The bill also makes data centers pay for the power they use.

This should accelerate the country the build more solar, wind, and battery projects.

https://www.npr.org/2026/11/20/g-s1-154579/senate-bipartisan...

[3 hidden]5 mins ago
> But major environmental groups, including the Sierra Club and National Wildlife Federation, have opposed the bill, arguing it will weaken environmental accountability.

> Many Democrats have also opposed the bill. They say it will worsen pollution by making it easier to greenlight oil and gas projects and energy-hungry data centers, while weakening protections for clean air and water.

[3 hidden]5 mins ago
Sierra Club is against building infrastructure to distribute clean energy as well. Their opposition to this doesn't say what you want it to say
[3 hidden]5 mins ago
The Sierra Club is not honest or trustworthy.

In 2020, they sued to stop a transmission line from bringing clean hydropower from Canada to Maine. In 2024, they sued the Maine government for not doing enough to combat climate change.

[3 hidden]5 mins ago
Oh yeah, how dare they sue to stop a private company from cutting a large corridor through pristine wilderness (maine's only really durable resource) to take hydropower from the people Maine buys it from to.... Not Maine.

Also Mainers get nothing out of it.

That transmission line was absurd and stupid. Maine needs new generation, not more demand to the same international markets we buy our power from.

Maine hasn't done enough to combat climate change, which is a serious problem for our fishing industry, which is already suffering from the impact of climate change and may eventually have no jobs. The power line made 0 improvement to that.

CMP has been refusing to connect already built solar installations. Our power is expensive because of that.

[3 hidden]5 mins ago
Is climate change an existential threat or not? Because if it's an existential threat I don't really see myself caring about a small amount of maine's pristine wilderness.
[3 hidden]5 mins ago
It's become necessary to destroy the environment in order to save it.
[3 hidden]5 mins ago
> to take hydropower from the people Maine buys it from to.... Not Maine.

Interesting attitude. Why should anyone in North America care about the effects of climate change if the worst effects are felt in... Not North America?

[3 hidden]5 mins ago
> making it easier to greenlight oil and gas projects

Orthogonal problem that ought to solved separately.

It would be weird for anyone to argue that we do not need more energy faster and at lower prices.

Energy is foundational to civilization.

[3 hidden]5 mins ago
No one disagrees that energy is foundational to civilization, but we have been thoroughly informed of the dangers of oil and gas consumption. This comes down to short term financial gains over long term environmental impacts, and AI companies are slated to consume unprecedented amounts of energy.
[3 hidden]5 mins ago
Gas turbines are not more economical than solar and wind. It's infuriating that we're allowing them to be built, but as time goes on and more solar and wind join the grid these plants will be relegated to peaker work instead of running full-tilt.
[3 hidden]5 mins ago
> Gas turbines are not more economical than solar and wind.

I would expect that companies would build the most economical power source that they could. If gas turbines are less economical, it seems strange that they are being built; there must be something more nuanced to that story (or companies that are building them see the economics differently somehow).

[3 hidden]5 mins ago
"Companies are efficient" is a lie. They will make decisions which make no sense all the time for short term gains
[3 hidden]5 mins ago
Ummmmmm...

"Economical" has different meanings depending on whether your time horizon is the next quarter or the next century. It also depends on whether you just consider your own pockets or those of the whole humanity. A lot of stuff, like the entire fossil fuel industry, is economical as long as Someone Else pays conveniently pays your debts! In any case, "economical", whatever it means, is not what we need. Sustainable, including economically sustainable, is.

[3 hidden]5 mins ago
There are issues other than cost, like speed and expediency. Solar and wind are far cheaper than gas and coal, but they take much longer and are harder to permit. Basically oil, gas and coal are favored by regulation and subsidy.
[3 hidden]5 mins ago
> I would expect that companies would build the most economical power source that they could.

Please stop expecting this, or at least take the time to figure out whether your priors are justified or not, because it sure seems like the facts aren’t supporting your world view these days.

[3 hidden]5 mins ago
I bet that in the US that manufacturing and installing a gas turbine is in order of magnitude less expensive than manufacturing and installing the equivalent solar capacity.
[3 hidden]5 mins ago
There is no equivalent solar without at least overnight hours of batteries -- which are, yes, very expensive.
[3 hidden]5 mins ago
Not true at all. You can order huge amounts of solar panels for the cost of a gas turbine with delivery in a few weeks, while you can't get delivery of turbines for several years [], but getting land and a permit is a different issue. You missed the real issue which is regulation. Meanwhile the likes of xAI data centers pollute with gay abandon.

[] https://www.nytimes.com/2026/11/03/climate/jet-engines-ai-da... https://www.wsj.com/business/energy-oil/how-jet-engines-are-...

[3 hidden]5 mins ago
How are people still saying this? You can spin down the gas turbine when you have a lot of solar production.
[3 hidden]5 mins ago
Why don’t they just build solar / wind / batteries if they’re so much more economical?
[3 hidden]5 mins ago
Because current regulatory and administrative barriers prevent them from being built or connected to the grid.
[3 hidden]5 mins ago
The point of life is to be healthy and happy. A substantial fraction of the energy we use actually has the opposite effect. We don't need more energy overall when we could use less with better life outcomes.
[3 hidden]5 mins ago
More energy in the form of fossil fuel-based sources is not cheaper, no matter how low the immediate price. The long-term cost of continuing to increase greenhouse emissions are exponentially larger.
[3 hidden]5 mins ago
It isn't lower on paper, either. BAAJA would unlock wind and solar projects moreso than any fossil fuel generation.
[3 hidden]5 mins ago
It's pretty clear that across the board leadership at Google, Meta, Amazon, Anthropic, OpenAI, Microsoft and others don't believe this.
[3 hidden]5 mins ago
Because the stock market does not incentivize longer term thinking.
[3 hidden]5 mins ago
Yeah, and it's pretty obvious the US is still quite primitive and has a lot of civilisation founding left to do.
[3 hidden]5 mins ago
> Yeah, and it's pretty obvious the US is still quite primitive and has a lot of civilisation founding left to do.

Tell us more.

[3 hidden]5 mins ago
It's important to point out that these Groups' claims are wrong.
[3 hidden]5 mins ago
> The bill also makes data centers pay for the power they use

Data centers paid for power they use. It's amazing how far the data center conversation has drifted from reality.

Data centers, like all large scale grid-connected projects, are also assessed fees for the infrastructure costs they incur.

The debates have been around additional math beyond that where some people think that additional infrastructure buildouts should be billed even more to data centers, beyond what is assessed for their own connection. The argument was that utilities are sizing infrastructure projects based on projections of more data center growth and it's not popular to bill infrastructure growth to anyone but data centers right now.

[3 hidden]5 mins ago
They have to pay for the power they use? What a success. That bill must have teeth.
[3 hidden]5 mins ago
do datacenters not pay for the power they use?

i'm not stupid, i know what you mean, but why say stuff this wrong way?

there are no new datacenters in San Francisco, yet my electricity rates have nearly doubled in six years. can we pass a law that says, my rates are frozen in 2019 too? it would benefit a lot more people.

the hard part isn't deciding, oh, use solar. the hard part is, reining in voters and tulipmanias. some voters like tulipmania, next week, we'll learn the hard way that "datacenter bad" isn't the win people think it is.

[3 hidden]5 mins ago
Let's talk about solar.

Median days from application to permit issuance in San Diego is 0 days while it is 4 days in NYC. In San Francisco it takes 42 days!

Permitting is only part of the problem. Guess how long it takes from signing a contract to switching on your solar system? In Nevada it is just under 60 days while in California it is ~150 days. Texas isn't much better, at ~120 days.

Bureaucracy literally saps energy.

[3 hidden]5 mins ago
there are a bajillion places around the world with no permitting. where are their big solar projects? i agree that permitting is a core issue, but how do you justify, "people can do whatever they want with their land" when you use it for solar, but not datacenters? everyone's income is someone else's expense, so it's not so simple, there's no objective or rationalist or effective whateverist or non-political answer to this problem. it's all political. the bureaucracy is a symptom of a popular political goal, and then when we look deeply into whatever technocratic, unenactable policies about energy, like "nuclear" or "solar" or whatever, we find it's still vibes at the bottom.
[3 hidden]5 mins ago
The main goal of the permitting process seems to be raising costs and slowing projects down. If the permitting were focused around encouraging renewable energy projects and keeping things safe while hampering fossil fuels, I would be all about it.

Making it harder to build makes rent-seeking more profitable, so slowing things down is a goal unto itself. It's not just vibes - the vibes are how to obscure the goal.

[3 hidden]5 mins ago
> there are a bajillion places around the world with no permitting. where are their big solar projects?

Here is a is map of the largest solar projects & locations: https://cleanview.co/power-projects/operating/solar-farms/us

[3 hidden]5 mins ago
This is the US, the same place we just talked about
[3 hidden]5 mins ago
China burns more coal than the US plus Europe plus the rest of the world combined.[1] No amount of regulation will make the slightest difference at the global scale. If you really care about pollution, what you should defend is severe sanctions on China and accept the cost this has. But of course, people don't really think it's worth the cost when it comes down to it.

[1] https://ourworldindata.org/data-insights/almost-all-the-grow...

[3 hidden]5 mins ago
China still emits about half of the CO2 per capita compared to the US and only just overtook the EU. Everyone needs to do their part proportionate to their population.

https://ourworldindata.org/grapher/consumption-co2-per-capit...

[3 hidden]5 mins ago
I think your point and the point you are replying to are consistent.

Your link takes a lot of the coal power CO2 that is generated in China and assigns it to the US where the products are consumed. The prior post argued that the US needs to stop importing products from China.

[3 hidden]5 mins ago
Here's emissions per capita without reallocation by consumption: https://ourworldindata.org/grapher/co-emissions-per-capita

It shows the same thing. US emits way more per capita than China.

[3 hidden]5 mins ago
I stand corrected.
[3 hidden]5 mins ago
Why cherry pick? Put Singapore in the check list.

If you care about CO2 per country, it's China. If you care about CO2 per person, there are many countries worse than the US.

https://ourworldindata.org/grapher/consumption-co2-per-capit...

[3 hidden]5 mins ago
I picked those because those are what the original comment mentioned. I just don't think it's fair to point fingers at China just because they have a lot of people when other countries are more wasteful.
[3 hidden]5 mins ago
>I just don't think it's fair to point fingers at China just because they have a lot of people

oh okay, why not?

population growth isn't just some random function.

brutal, maybe -- but unfair? Not sure.

[3 hidden]5 mins ago
You could go by land area I suppose?
[3 hidden]5 mins ago
US: ya but China pollutes more!

China: ya but the US are far more dirty per person they'll gain an advantage this way so we can't slow down (though they seem to be the ones building out the most renewable infrastructure)..

Canada: ya but both those guys are worse than us. We barely have any people!

India: well if nobody else even nice guy Canada is going to bother...

It's like nobody remembered the eye for an eye teaching.

[3 hidden]5 mins ago
This isn't about punishment, it's about collective action to eliminate fossil fuel usage, and looking at individual consumption is the right way to look at it. You imagine the average Chinese person living in a cold climate can't heat their home, and wants to increase their heating usage by 200kg per year. Meanwhile the average person in the US uses 700kg for heating and also spends a few tons on flights and car travel every year.

It's not about "advantage" it's about asking the Chinese person to continue to freeze while the person in the US reduces their consumption by 200kg. So the Chinese person has 200kg worth of heating while the American has 500kg. But they should be equal assuming equal climate, is the thing.

[3 hidden]5 mins ago
well American need more because they drive 2 ton ford car
[3 hidden]5 mins ago
"The elves should take the ring" "No, the men should!" "No, the dwarfs should take it to Mordor"

The hobbit steps forward

"I won't. I won't take it. We hobbits are too small, we never got a ring of power. We have no involvement in this. It's simply not out responsibility!"

The men, elves and dwarfs, not humiliated, continued to argue.

[3 hidden]5 mins ago
By that logic if you split China into ten countries with the same total emissions then they wouldn't matter any more right?
[3 hidden]5 mins ago
CO2 is not the worst effect from coal
[3 hidden]5 mins ago
The climate doesn't care about per capita.
[3 hidden]5 mins ago
Thought-terminating nonsense. All those people you call "capita", are real people with real lives and your suggestion in essence is to immiserate them. The climate does care about per capita when each of those people contributes to the development of green technology even in some indirect way.
[3 hidden]5 mins ago
> All those people you call "capita", are real people with real lives and your suggestion in essence is to immiserate them.

You're putting a lot of bad faith words in my mouth I never said. If anything is "thought-terminating," it's that.

[3 hidden]5 mins ago
You should know that the phrase you used implicates the beliefs I outlined, if you didn't already.

Also I offered a real counter-argument, that per-capita is in fact important. How is this thought-terminating?

[3 hidden]5 mins ago
the climate does not care about per capita!
[3 hidden]5 mins ago
Correct: the climate cares about absolute numbers. If we want to distribute an absolute reduction in carbon fairly across the world, we should equalize the per-capita emissions.
[3 hidden]5 mins ago
Nobody cares about fairness; in order to convince larger nations to do anything , you need a large enough military and economic force backing the alternative.

I'd actually say that the US AND China have little-to-no responsibility to start a push towards cleaner energy because they benefit so strongly from the status quo. Those that don't have that benefit should be more incentivized to pushing their carbon output to zero, and pressuring both these entities with sanctions, trade wars, and military action wherever possible. Eco-terrorism should also be financed by these smaller entities targeting backers and companies of the larger ones.

[3 hidden]5 mins ago
Fuck it. Let's just restart the game
[3 hidden]5 mins ago
It also doesn't care about what countries people are arbitrarily grouped into. Per person emissions need to be reduced, and bigger countries have more people. You can't blame countries for having more people if their emissions for those people are lower than your own.
[3 hidden]5 mins ago
sure i can. the important counts are the amount of plants and cars and so on.

the thing that needs to be turned off is coal and gas plants, not to mention gas and diesel cars and trucks

the people are irrelevant, and this is just excusing larger and growing populations to burn more.

i can blame them just fine for having more and replacing their coal plants with new coal plants

[3 hidden]5 mins ago
This (climate) is a common problem. You of course can say "I want my yacht and my personal airplane and I do not care if 1M in another country use a coal plant only to have electricity to cook a meal". But I hope you can imagine how those 1M will consider you.

Solving a common problem can require finding a common solution, which would require some sort of agreement like "I will consume less, if you consume less and we can both survive decently".

Anyhow, at this point I find the discussion irrelevant as renewables are cheaper. Whoever wants to make it about "save the planet" or "they and not us are doing the damage" are indirectly helping the fossil fuel industry that wants this to be a "culture war" and not a discussion about "what is cheaper" (which they would loose).

[3 hidden]5 mins ago
>the people are irrelevant

That’s where you lost the point. Earth doesn’t care if it gets hotter. The people do.

[3 hidden]5 mins ago
Right. In that case China+US (which is strictly bigger than either) is the singular entity which needs to regulate.
[3 hidden]5 mins ago
the climate doesn't care about china or europe or america either, so why are we apportioning blame to individual political entities in the first place? and why are you chiming in with this perspective after it's pointed out that, relative to its population, china is less polluting than america?
[3 hidden]5 mins ago
> per capita

So? If I have a family 5 times the size of your family, my family is consuming more, negatively affecting those around me.

[3 hidden]5 mins ago
Let's say there's your big family and five other families who all drive big trucks. The other families are all pointing at you saying you're the one who needs to fix things, not them, even though their collective emissions are higher.

It doesn't matter to the planet what arbitrary social construct you group people into, every family needs to put in the same work per person.

[3 hidden]5 mins ago
> every family needs to put in the same work per person.

Why?

[3 hidden]5 mins ago
Because the people are the ones who pollute, not abstract concepts like countries, states, families, etc.
[3 hidden]5 mins ago
People's individual choices make zero dent in the overall environment, even if it affects local ecology.

Environmental issues have to be solved at a group-scale. Policy, market forces, technology innovation, etc.

To tell people they need to stop using plastic, not get something delivered next day, to not eat meat, and make it their personal responsibility simply does not solve things.

What's really off-putting (and likely hypocritical) is someone telling someone else they're failing morally on these matters.

[3 hidden]5 mins ago
My point is that the groups are arbitrary. If you have a country with a billion people or ten countries with a hundred million people then the responsibility is exactly the same. Splitting a group into smaller ones doesn't make it not matter any more.

And of course personal choices matter, exactly for this reason. If a million people choose to be wasteful then it's not valid to claim "it's not MY massive car that's the problem".

[3 hidden]5 mins ago
China chooses to manufacture disposable, low quality crap, and then ship it around the world. So casting them as better because of low emissions per capita is not holistic thinking.
[3 hidden]5 mins ago
They manufacture it because people around the world buy it, it's not like they do it to hurt the environment like a Captain Planet villain.
[3 hidden]5 mins ago
The ones to blame about the disposable crap are the ones that buy them anyhow. Which is mostly not China, so either way they seem to be "less to blame".

Or, by that logic, let's go to the source and blame only the fossil fuel producers ... wonder who produces more oil.

[3 hidden]5 mins ago
If you are going to blame consumers as the ultimate drivers (e.g. in the case of goods manufactured by China) then you also have to blame them for the drive of other things like fossil fuels or AI data centers.

Otherwise it is just cherrypicking.

[3 hidden]5 mins ago
Well obviously, if you burn the petrol in your car it's not the gas station's fault for selling it to you.
[3 hidden]5 mins ago
The point I was trying to make is that it makes little sense to point to one type of link in the chain (material producer/energy user/product consumer), everybody gets something out of it.

Now, how essential for them is what do they get out of it - that's a different story. Buying one mobile phone per year is different to not having electricity to turn on the lights at night (exaggerated examples).

[3 hidden]5 mins ago
And if you have a common last name, such as Smith or Johnson, people with your last name are consuming several orders of magnitude more than people whose last name is Zillman.

The environment doesn't care about arbitrary sorting of humans into sets. It just cares about totals, which is why unless you can make a good case that some people have some sort of natural or moral or divine right to consume or emit more than others the allocation has to start with per person.

[3 hidden]5 mins ago
> It just cares about totals

Global greenhouse gas emissions have risen sharply over the past century, driven by industrial growth, expanding energy use, and population growth.

Select for US and China in this chart - https://ourworldindata.org/greenhouse-gas-emissions

What do you recommend?

[3 hidden]5 mins ago
by doing nothing, most of western country birthrate is in decline anyway

would take a while for developing country but at the end of times, world population would decline and in translate to "reset"

[3 hidden]5 mins ago
The problem of per capita has a very simple solution. Those with too high CO2 per capita need to have more children, and those with too low CO2 per capita need to have fewer children. That way it becomes more equal and thus more fair.

Should this be the new goal for COP32, or should it focus more on addressing global climate change and reducing the total global emissions? What is the metric that we actually care about?

It should be noted that older demographics have higher emissions rates than younger ones, so having more children is really an effective way to reduce per capita CO2 emissions. The emissions get spread out over larger population, and the new generation has less emissions than then the older generations.

[3 hidden]5 mins ago
As soon as a metric becomes a target it ceases to be a good metric, obviously. The point of per capita emissions is looking at things like heating. If 1M people living in a cold climate in China use 1000 tons/year for heating, and 100k people living in a similar climate in the US also use 1000 tons/year for heating, expecting both countries to cut those emissions by 500 tons/year is likely suggesting that the Chinese should literally freeze to death.

Serious proposals need to take into account individual needs.

[3 hidden]5 mins ago
Conversely China isnt setting up the sites and turbines/coal in almost residential areas. With the accompanying noise/dust pollution and other hazards , all in the pursuit of makinga quick turnaround instead of following the existing regs for greenfield projects.

China projects as far as i can tell are in remote , renewable heavy areas and mostly greenfield with the coal/gas contribution being mostly to stabilise the grid (when it not in industila/manufacturing processes).

And again this is a country of 1.4B people that also increasingly the manufacturing base for most of the planet, so throw in all the goods going to SouthAmerica/Africa/Europe into that energy consumption calculation. That lage parts of the developing and develpoed countires can accelearte their solar/wind buildouts comes directly from those coal/gas plants in Chna.

[3 hidden]5 mins ago
IIRC China also generates more solar power than the rest of the world combined. It's just a really big place.

It's like saying the USA has the most inmates out of any country. True but not surprising because it's the 3rd biggest and has a crime problem. The USA has more inmates per capita, now that's the real problem.

[3 hidden]5 mins ago
Not necessarily. Countries like China tend to execute entire classes of prisoners resulting in a lower per capita incarceration rate. Some countries are highly corrupt and laws are not enforced as they should be and would have similar rates if they were. How do you determine the 'correct' rate?
[3 hidden]5 mins ago
“Mercy to the wolves is cruelty to the sheep.”
[3 hidden]5 mins ago
Most estimates put China at ~1000-2000 executions per year.

The US incarcerates 1.2 million people.

[3 hidden]5 mins ago
China has almost half the CO2 emissions per capita compared to the US, and only surpassed the EU as recently as 2019 – and that's even a biased analysis considering the worldwide industry is there.

https://ourworldindata.org/grapher/co-emissions-per-capita?c...

[3 hidden]5 mins ago
the climate does not care about per capita!
[3 hidden]5 mins ago
By this reasoning, if China split into a hundred tiny countries tomorrow, the problem would be fixed, since each individual micro-China wouldn't be a big contributor.
[3 hidden]5 mins ago
that made me laugh
[3 hidden]5 mins ago
Such a lazy argument, climate does not care about political boundaries.
[3 hidden]5 mins ago
the climate doesn't care about anything, it doesn't have the capacity to do so. The capita though...they can do some caring.
[3 hidden]5 mins ago
Shut up. You are expecting some koombaya shit where everyone in the world agrees you are right and decides on a maximum emission per person and implements it. That's not how anything works.

The largest entity that can agree and implement anything like this across its parts is a country, so you have to appeal to countries to implement regulations.

[3 hidden]5 mins ago
Degrowth mindset. China's trajectory is lower emissions per capita and diversification of power, and global exports of solar technology.
[3 hidden]5 mins ago
I don't know about statistics for the entire country but the two most populous states in the country have been growing their percentage of solar power at high rates.

This surge of fossil fuel use is only possible because we have a glut of fossil fuel plants that were sitting idle because of solar.

[3 hidden]5 mins ago
Regulation obviously could approach this issue, look at China's record on the Paris agreement and their track record inside the WTO. Regulations will obviously be able to effect the incentives that shape their economic policy. Additionally, China's coal consumption of 3.4t/capita doesn't differ that much from the United State's coal consumption from 1985 to 2010, which was above 3.0t/capita for the entire duration of that period, and Australia's current consumption of 3.4t/capita. Are you going to sanction Australia also?
[3 hidden]5 mins ago
Europe is implementing a tariff on imports that is scaled based on the carbon footprint of the item. In that way aluminum produced in Iceland by clean electricity would be more attractive on price than it is today than aluminum produced in China from a heavy coal grid.
[3 hidden]5 mins ago
China's coal consumption is irrelevant to whether a US data centre should be required to control its local pollution.
[3 hidden]5 mins ago
I don't see the point in bringing up China and only gauging by population alone for amount of energy used. That completely ignores factors like resources access, infrastructure, geography, and other factors I'm sure I'm missing in the energy equation.
[3 hidden]5 mins ago
Green energy activists tend to present moral arguments which are easily dismissed on economic grounds. Even if the US were 100% green energy, it makes no meaningful difference since the rest of the world still generates most of the pollution. That fact is both inconvenient and inescapable—it has been very counterproductive for green activists to dance around it by citing unrelated data as a rhetorical device to change the frame.

This new wave of green activists have learned to simply make economic argument instead. To their surprise "it's cheaper and it's clean" sells a lot better than 'non-green energy is h!tler and we're cheaper." I also think markets needed time to get comfortable with green energy.

[3 hidden]5 mins ago
> Even if the US were 100% green energy, it makes no meaningful difference since the rest of the world still generates most of the pollution.

For a rough calculation, if the US had emitted no CO2 this millenium, then if all else were equal that would be more than 100 gigatonnes less CO2; meaning we'd be closer to 420ppm than the current 432ppm.

(There are feedback loops like oceans and plants that would absorb less CO2 at lower concentrations which I don't know how to account for.)

Mass of CO2 and ppm: https://en.wikipedia.org/wiki/Carbon_dioxide_in_the_atmosphe...

[3 hidden]5 mins ago
Yes it will and to make global treaties on emissions rampdown politically realistic, people need to get the ducks lined up at home in each country.

It's a classic collective action problem, and the international community has done similar things before.

[3 hidden]5 mins ago
You need a global carbon tax and China has to implement it or sanctions. They'd probably go with the tax as the biggest producer of solar.
[3 hidden]5 mins ago
China is not going to get sanctioned.
[3 hidden]5 mins ago
Europe is implementing a carbon price on imports.
[3 hidden]5 mins ago
It's that or eco-terrorism sponsored by smaller nations (or war). If we don't get any of those, then we're not going to solve the climate crises and we're all going to bake ¯\_(ツ)_/¯
[3 hidden]5 mins ago
How is severe sanctions on China supposed to solve anything? How would you convince a country you just sanctioned to do what you want (invest in green energy)? Did that work out with any other country we sanctioned??

China is also #1 when it comes to green energy research, and guess what, they got there by investing their dirty fossil fuel profits.

If we care about pollution we need a massive global Green New Deal. Yet all we get from Europe is lip service about it, in America we are actively kneecapping renewables, and only in places like Australia do we see some success.

The only superpower investing on a continent scale in electrification and renewables is China.

[3 hidden]5 mins ago
Chinese coal and CO2 emissions are dropping https://www.carbonbrief.org/analysis-chinas-quarterly-co2-fa...

A carbon tariff on China is an excellent idea. As long as that same tariff applies to the United States.

[3 hidden]5 mins ago
Likewise, the rest of the world combined pollutes far more than China, so no amount of sanctions against China will make the slightest difference.

We have all the excuses we need to do nothing.

[3 hidden]5 mins ago
The world turns a blind eye to everything China does so they can manufacture trinkets for us that can be marked up by 90%.
[3 hidden]5 mins ago
We should do that too.
[3 hidden]5 mins ago
You never have to look far to find the predictable "but China!!!!" comments. So:

1. China has 4 times the number of people as the US;

2. China is using coal in part because of the disruption to oil and natural gas supply caused by the US boondoggle and grand misadventure in Iran. As an aside, China essentially saved the world economy by cutting their oil imports by so much;

3. China has built in recent years and continues to build more renewable energy (particularly solar) than the rest of the world combined;

4. For pretty much anyone else building renewables, it's China supply the tech more often than not;

5. China has built the world's largest hydroelectric dam (ie the Three Gorges Dam) and in the coming years is building something three times as large;

6. China is actually building nuclear power plants, something HNers normally love;

7. Since the 2020 OPEC oil supply shock (which never gets blamed on the Trump admin for some reason), China has added roughly 80 million EVs to the road. It's estimated that this has reduced global oil consumption by approximately 2 million barrels per day, which again has greatly contributed to saving the world economy but also represents a ~2% reduction in global oil demand.

What else could China do? What is anyone else doing in comparison?

[3 hidden]5 mins ago
Also Wikipedia claims that China plants lot of trees (reforestation project).
[3 hidden]5 mins ago
china could not try to manufacture too much of absolutely everything, at which they wouldnt need to continue producing so much fossil fuel based energy.

theyre purposefully wrecking as much as they can for industrial independence and dominance. thats a bad choice for everyone, but especially the environment

[3 hidden]5 mins ago
China is already having a manufacture flight as cost of living goes up and companies look into cheaper neighbors like Vietnam.

China could do more, but there comes a point when its time to stop pointing at the place making real inroads into alternative energy and turn to the ones screaming drill baby drill with a skitzo energy policy.

[3 hidden]5 mins ago
I'm with you on the futility of trying to solve global carbon emissions with even more regulatory burden in first world countries (which have already had declining carbon emissions for decades, while the rest of the world's emissions skyrocket), but what leads you to believe a confrontational, sanctions-heavy approach to Chinese carbon emissions is the only viable approach to Chinese carbon emissions?
[3 hidden]5 mins ago
Fixing these cost calculations is what a carbon tax does. I bet a carbon tax on data centers would be popular.
[3 hidden]5 mins ago
> popular

First you'd have to explain to things to the populace:

a) how those taxes will affect their purse

b) what those tax dollars will be used for

I bet you that on (a) the cost will be passed on (because we are the ultimate consumers) and on (b) the tax dollars will be misspent to a meaningful degree. End result is not a meaningful improvement to reduced environmental impact and consumers pay more. Basically wealth evaporating.

[3 hidden]5 mins ago
> (a) the cost will be passed on (because we are the ultimate consumers)

Tax Incidence is a lot more complicated than that

[3 hidden]5 mins ago
It doesn't matter if the tax dollars are literally burned on receipt, a carbon tax would cause a meaningful improvement to environmental impact.
[3 hidden]5 mins ago
> I bet a carbon tax on data centers would be popular.

Shouldn't a carbon tax apply everywhere, equally?

[3 hidden]5 mins ago
Maybe but that would be unpopular.
[3 hidden]5 mins ago
I think we can start with the ones who have the money to pay
[3 hidden]5 mins ago
A carbon tax will not get one republican vote in the Senate. And it needs a dozen to override the veto.
[3 hidden]5 mins ago
The closure of the Straight of Hormuz and subsequent Democratic wave shows the opposite.
[3 hidden]5 mins ago
How do you figure that either of those says that a carbon tax on data centers would not be popular?
[3 hidden]5 mins ago
.. on data centers, not on the general public.
[3 hidden]5 mins ago
This would be a perfect way to start a carbon tax in the US, considering how widely public opinion is against data centers.

Plus it would have the net effect of incentivizing renewable power to them, so all the assets aren't dark for years when the bubble pops and the economics turn against fueling them with gas.

[3 hidden]5 mins ago
do you think concentrated feeding operations could survive a carbon tax?
[3 hidden]5 mins ago
That's why we should force companies to report on SASB metrics from their inception. A lot of these "oh look how green we are" reporting is also heavily manipulated to make it look good only on the surface.
[3 hidden]5 mins ago
I think this is better thought of as a failure of public utilities and regulation itself. I'd bet that these small scale power installs are more expensive per kw than a proper grid connection. We've been investing massively in data centers since 2022. The utilities didn't plan ahead for growth. They should have used the future revenues to finance cleaner power plants. But they are too slow and overly regulated in all the wrong ways.
[3 hidden]5 mins ago
Maybe black holes are just the remnants of civilizations who discovered self improving agi...and then consumed all their energy into a singularity, while trying to prove their ultimate, most interesting, math problem.
[3 hidden]5 mins ago
It's all paperclips in the end.
[3 hidden]5 mins ago
Tears in the rain

Paperclips in the wind

Keep optimizing until GOTO END

[3 hidden]5 mins ago
I would expect most of the compute to be dedicated to generating reports, reading emails, bots arguing online, and other non productive activities
[3 hidden]5 mins ago
It's more likely they poured it into bitcoin mining.
[3 hidden]5 mins ago
The math problem was: "what is 6 times 7" ?
[3 hidden]5 mins ago
It's why we need smart regulation, like a carbon tax. Our currently regulatory structure massively favors fossil fuels.
[3 hidden]5 mins ago
The problem isn't not knowing what to do, the will is simply not there. That's the saddest part.
[3 hidden]5 mins ago
The will to do what? Regulate? I would disagree.
[3 hidden]5 mins ago
Tax companies 1% of their revenue. Allow for deduction of batteries, renewables or charging stations only for the next 10 years.

Swap tarrifs for carbon tax on imports.

[3 hidden]5 mins ago
Naively, taxing revenue at 1% would seemingly make it impossible to run any volume-based cashflow business with a <1% margin on COGS, e.g. many grocery stores.

That being said, as it happens, those businesses also happen to have heavy logistics involvement, so in theory, EV tax credits could save that business model. Though that "logistics involvement" is currently arm's-length (they contract with trucking companies) so in order to benefit from those credits by forcing EV fleet conversion, they would need to vertically consolidate, i.e. buy up all those trucking companies.

And of course, under your model, after 10 years, they'd lose these credits and become negative-margin and cashflow-starved anyway.

[3 hidden]5 mins ago
> Naively, taxing revenue at 1% would seemingly make it impossible to run any volume-based cashflow business with a <1% margin on COGS, e.g. many grocery stores.

it will be factored into price for end consumers and back pressured to suppliers

[3 hidden]5 mins ago
IOW, it will have 0 impact on Anthropic with its 80% profit margin, but will make groceries more expensive.
[3 hidden]5 mins ago
I just explained clear error in your speculations: groceries won't go out of business.

Impact is separate topic of discussion.

[3 hidden]5 mins ago
"back pressured to suppliers?" Just like "other countries" paid Trump's tariffs, not American consumers. LOL.
[3 hidden]5 mins ago
> Just like "other countries" paid Trump's tariffs

You obviously don't have any numbers in your hands about this.

[3 hidden]5 mins ago
pot. kettle. black.
[3 hidden]5 mins ago
burden of proof
[3 hidden]5 mins ago
On the person who made the first assertion. You, not me.
[3 hidden]5 mins ago
I don't agree with "first" criteria. I would say "first easily challengeable assertion" or "assertion which is currently under discussion", and this part is on you.
[3 hidden]5 mins ago
Naive, indeed. In reality all grocery stores raise prices by 1%, which is like 1 month of price inflation.

The argument against taxing revenue is usually framed in terms of "thin" businesses that perform some administrative function but not much of a physical one, not businesses with significant operations that still happen to be in a mostly-competitive market. Although perhaps you could argue that this competition is merely a skin over the real pricing power (eg egg cartel, PG vs Unilever), and grocery stores have become akin to thin businesses.

Then again, individuals are essentially taxed on revenue (can't deduct most things that would be straightforward business expenses, like "commuting" or food), and we still continue to exist (at least for now).

[3 hidden]5 mins ago
No economic policy which can be explained in one line will ever work.
[3 hidden]5 mins ago
We already tax at rates much higher than 1%, and already allow substantial credits and deductions for solar, renewables, and biofuels.
[3 hidden]5 mins ago
The tax is on profit, not revenue and Anthropic runs at a loss.
[3 hidden]5 mins ago
But taxes on loss is paid by someone else in the end. Nvidia or the company which provides power infrastructure(and the people who work on them) has to pay the tax.

Why not just tax the use of gas and coal directly, rather than going through all the loop.

[3 hidden]5 mins ago
Everyone going on about Elon, Beniof, etc, shifting their tune when Republicans are in, when Dems are in

Until 5 years ago rank and file among IT were anti-union, regulations... now they're the frogs feeling the heat and it's all "regulate everything"

Not one mfer is socially reliable. Everyone really should just be minimized to a single individual, a single vote and left to their childhood neighborhood.

It's all such obvious copy-paste and empty platitude effort from top to bottom with humans.

Mirror neurons most def seem like the simplest answer; superficial monkey see-monkey do follow along without critical thought happy paths in the brain being lit.

Fingers crossed the AI bubble pops and ends it all one last gasp, our answer to the Fermi paradox. Or at least makes everyone obsolete and ignorable.

[3 hidden]5 mins ago
A plain old wellhead tax solves this and every other problem in this category.
[3 hidden]5 mins ago
is that a tax of all gas or a gas on the gas that Anthropic is using in this scenario?
[3 hidden]5 mins ago
As the name suggests it is a tax on taking the stuff out of the ground in the first place.
[3 hidden]5 mins ago
The scale of energy that AI data centres use is huge in a way that most people don't have the ability to visualize. They think of a computer that their cat likes to sit on and multiply that gentle warmth by some number in their head. This is completely unrealistic.

Here's a quote about a data centre currently being built near Edmonton, Alberta.

------

"Utah State University physicist Robert Davies was hired by SLCN to conduct a preliminary technical assessment. He said if the Wonder Valley project goes ahead it would release the equivalent of 23 atomic bombs of energy every day, though he noted the comparison has its limits."

Source: https://www.cbc.ca/news/canada/edmonton/wonder-valley-slcn-c...

------

This is not the total energy being consumed by this data centre. It's merely the waste heat it will release to the surrounding environment.

Here's the energy consumption:

------

"At full operation, it would consume energy at a continuous rate of 18 gigawatts or about 1.5 petajoules per day and emit about 30 million tonnes of carbon dioxide a year – more than four per cent of Canada’s total emissions, said Davies, whose assessment was based on publicly available information."

------

That's the equivalent of 1.66 million Canadians in carbon emissions, or 1.6 Edmontons. This is just one data centre. Many more are going up across Canada, and Canada is both late and slow to the buildout.

That's the kind of scale we're dealing with at a time when, as a world, we are trying to reduce our carbon emissions.

[3 hidden]5 mins ago
> at a time when, as a world, we are trying to reduce our carbon emissions.

From my perspective, that is not quite the correct way to put it. Because honestly we are not even trying.

---

I find those "Olympic swimming pool type" comparisons very tiring, to be honest. They certainly don’t help my grasp anything. "23 atomic bombs of energy" yeah okay

[3 hidden]5 mins ago
It sounds really dramatic. Even worse, unlike an olympic swimming pool, nuclear bombs vary drastically in terms of energy output. Like, several orders of magnitude. If they mean 18GW as mentioned later in the article, that's about the energy consumption rate (total, not just electricity) of a city like Houston.
[3 hidden]5 mins ago
if the prices of domestic and international flights were cheap, people would be flying a lot more, I saw a girl say stop using generative ai because it hurts the environment as a caption on her trip to Iceland, the irony is staggering.
[3 hidden]5 mins ago
The prices of flights are cheap.
[3 hidden]5 mins ago
One thing that make absolutely no fucking sense for me is that in places of Europe, for intra country (so good train network and dense population and no so long distance) train is quite more expensive that (subsidized) plane.
[3 hidden]5 mins ago
its the costs. Trains are heavily subsidised because they have to be.

1. Trains need tracks for the full length of the journey. This is a lot to maintain. Plans just need a runway each end and air in between. 2. They are slower so staff have to be paid for longer hours.

However, for many journeys trains can still work out cheaper because they can take you closer to where you want to go from closer to your starting point, so the cost of the complete journey can be lower.

[3 hidden]5 mins ago
Trains have to deal with thicker atmosphere (closer to the ground) and have tracks that need maintenance. Planes only need a few square miles of tarmac.
[3 hidden]5 mins ago
Why just Europe then.
[3 hidden]5 mins ago
Trains are also more expensive in the US
[3 hidden]5 mins ago
Why it doesn't. Running trains is simply expensive regardless of country. Swiss have excellent network and its very expensive to use even for locals (I live here and use it for daily commute), for tourists its prohibitive. Other places have often much less quality / reliability / reach and prices are still high.

Cheap train network is a proper pipe dream, not possible. Much better to have fully self driving electric taxi cars (so not that joke with driver constantly needing to be aware if your tesla wants to steer you to standing ambulance / firefighter trucks on highway, I mean no driver setup to push prices to very low levels).

[3 hidden]5 mins ago
domestic flights in the USA are up 25% year over year and international flight are up 15%, do you have early Anthropic equity?
[3 hidden]5 mins ago
This doesn’t change your argument but it’s worth pointing out:

> This is not the total energy being consumed by this data centre. It's merely the waste heat it will release to the surrounding environment.

Total energy should in theory equal waste heat. All energy a data center consumes will eventually get converted to heat.

[3 hidden]5 mins ago
Why not use heat from datacenters to heat homes? Or to produce more electricity? Or to heat greenhouses in winter? Looks like free energy.
[3 hidden]5 mins ago
> This is not the total energy being consumed by this data centre. It's merely the waste heat it will release to the surrounding environment.

What would happen to energy it uses other than be released as heat? It has to go somewhere (conservation of energy) and I cannot think of anything other than heat that energy used in a data centre would become.

> about 1.5 petajoules

That is about 2% of total Canadian energy usage. How many more are going to go up? Are they all going tbe powered predominantly by fossil fuels?

[3 hidden]5 mins ago
> What would happen to energy it uses other than be released as heat?

If only we had some need for heat as a civilisation and could actively combine projects to achieve multiple things at once.

[3 hidden]5 mins ago
It would still be released as heat.

It could reduce the use of other energy that is used JUST to create heat, so its more efficient - but that only works in some locations, and in most of those only for part of the year.

[3 hidden]5 mins ago
> This is not the total energy being consumed by this data centre. It's merely the waste heat it will release to the surrounding environment.

Is there a difference? Where else would the energy go?

[3 hidden]5 mins ago
I imagine there are thousands of optical transceivers, emitting milliwatts each. That adds up to literally dozens of watts!
[3 hidden]5 mins ago
>The scale of energy that AI data centres use is huge in a way that most people don't have the ability to visualize

While I can't really visualise energy at all, this does make it seem like more than the 0.5% of world electricity consumption that it actually is.

While it is currently cheaper to build renewable energy capacity than most fossil fuels, as far as I can tell, the reason most data centers are not doing so is because they can't, All of the renewable infrastructure being produced is currently being deployed.

The question they should be being asked is when will they be switching to renewables. The height of the plot is not the important thing. It is the area underneath that matters,

[3 hidden]5 mins ago
0.5% of the world's energy is massive! Every other thing society does - agriculture, shipping containers across the ocean, flying planes - and 1/200th is for AI? The thing we didn't have a couple years ago and we were all fine without?

"When are you going to switch to renewables?" has proven to be a useless question. Companies will say whatever is expedient, and then they'll buy credits or whatever green washing instead of actually reducing consumption. If you let them double their energy consumption by adding renewables they'll just use double.

[3 hidden]5 mins ago
For starters, 0.5% of the world's electricity, not energy.

Most of the worlds energy use is not electrical.

It is a large amount. I do not deny that, I just want people to consider it in terms that are true. Every person who I have asked how much they think it is based upon the coverage they have encountered has massively overestimated.

A common response is that considering the concern reported in media they estimate data centers use 10-15% of electricity. How can we have an informed conversation about the relative importance of things if people have been misled about the relative scale.

[3 hidden]5 mins ago
Under the rate of expansion we saw in 2025, data center share of electricity in the US is expected to be around 12% in 2030. So it does not feel particularly misleading

Source: https://eta.lbl.gov/publications/united-states-data-center-e...

[3 hidden]5 mins ago
On the other hand 2030 projections are for 1.5% of electricity globally.

The US figure is not so much a factor of data center usage but rather the desire to centralise contol within the US where energy providers are more interested in extracting profit from existing infrastructure than building new infrastructure.

Add to that the amount of electricity spent on industrial uses that the US just doesn't do any more. It inflates the US percentage.

That's not a symptom of data centers, The US is trying to put a new engine in a neglected car with a rusted out chassis.

[3 hidden]5 mins ago
The US has a particular high concentration of data centres.

Those are not just AI ones, and some reduce other energy usage - e.g. if a video conference replaces multiple people travelling to a face to face meeting.

[3 hidden]5 mins ago
But it's not huge. All US data centers consume about 5% of total US electricity. Projected to be 7-12% by 2028. Quoting numbers without context makes them seem much larger than they are.

https://www.energy.gov/articles/doe-releases-new-report-eval...

[3 hidden]5 mins ago
5% of total US consumption is a huge number though. More then doubling it in 2 years makes it even worse.
[3 hidden]5 mins ago
They have not broken ground yet on this project, it still has legal hurdles before building starts. The last 10 largest built datacenters don't even add up to this 9GW mega project estimate which wouldn't finish until the mid to late 2030s if it even gets built. U.S. total electricity growth per year seems to be 2-3 percent, if 9GW+ projects become the norm and are built at scale you will see this number skyrocket. Your comment led me to believe this is the norm and 9GW is / has already been built at scale.
[3 hidden]5 mins ago
> It's merely the waste heat it will release to the surrounding environment

Is it not possible to use some of that for basically free local district/water heating?

[3 hidden]5 mins ago
Totally possible, but just like a solar+battery setup, it costs more so why would they voluntarily do it?
[3 hidden]5 mins ago
datacenter builders have been very generous about giving things to the surroundings, and ultimately its another revenue stream
[3 hidden]5 mins ago
Infomaniak does this with Euria in Geneva, Switzerland.

https://news.infomaniak.com/en/euria-sovereign-ai-assistant/...

[3 hidden]5 mins ago
They’re _really_ noisy, so usually built far from the kinds of big cities that would benefit from district heating
[3 hidden]5 mins ago
People keep thinking that the solution is to just keep giving the thinking over to AI, and keep on waiting for it to grow.

Meanwhile codebases get exponentially more bloaty and buggy. The internet gets filled with more and more disinformation. Students get worse and worse.

And the growth everyone is depending on is also so cleary destructive. This is not sustainable and it will not be the solution.

[3 hidden]5 mins ago
Seems like the DC operators could buy some goodwill if they provided district heating?

Is that feasible at all?

[3 hidden]5 mins ago
> 23 atomic bombs of energy every day

What yield?

[3 hidden]5 mins ago
(18GW * 1 day)/23 ~= 67TJ

Little Boy (the type of bomb dropped on Hiroshima) was 63TJ

[3 hidden]5 mins ago
Where are you getting 18GW from? Last estimate I saw (for 2025) was 14.4 (125TWh continuous)

The IEA is projecting 500TWh (57GW ) by 2030 so it might be consistent with their figures.

[3 hidden]5 mins ago
From the source of the 23 atomic bombs claim that was linked in beloch's comment - https://www.cbc.ca/news/canada/edmonton/wonder-valley-slcn-c...
[3 hidden]5 mins ago
Ah, now I remember the original headline.

Now it has a bit at the bottom This story and its headline have been updated to provide further context about the preliminary nature of Robert Davies’ assessment and align with our journalistic standards and practices

I was thinking they were talking about current usage. Not guesses of how much something might use.

The only other source I found was https://majorprojects.alberta.ca/details/Wonder-Valley-AI-Da...

Indicating the initial build would be 1.4GW.

I'm not sure how much size matters of individual data centers apart from their contribution to the total. If they sum to the same amount, how much does it matter.

If it's true that you can build something of that size without an environmental impact report then there's something pretty wrong with the laws there.

[3 hidden]5 mins ago
Probably 50 kilo grams of TNT, affectionately known as "Fart Man".
[3 hidden]5 mins ago
I would note that this data center is likely unique among Canadian data center projects for being built to be an oil-and-gas consumer. They're presumably doing that with this Alberta data center in particular, because oil-and-gas is the natural resource Alberta happens to have, and it's strictly more economical to burn the extracted gas near where it's produced than to have to send it thousands of miles via pipeline first.

And that economy-of-locality means that if they wanted to build more of these gas-plant data centers, they'd just also build those in Alberta.

Thus, if they're building DCs "across Canada", then I would expect that the data centers being built in the rest of Canada will not be powered with piped-in Alberta oil. A data center built in e.g. British Columbia / Ontario / etc, would presumably use the natural resource common to those provinces, i.e. hydro power.

---

Also, coincidentally, there's something interesting/unique about building data centers in Alberta, that being that it's friggin' cold there. Which means that Albertans spend a lot of grid power (or oil-and-gas directly) to heat their homes each year.

They're already doing waste-heat recovery district heating for data center projects in Finland. Could be done in Alberta too. (Though maybe it'd be better to target some of these DC projects to slightly northern-er parts of Alberta, where it stays cold for more of the year, so that ideally the district heating is a desirable consumable year-round rather than only for 6 months out of the year.)

---

As for carbon emissions, I would point out something less positive and more frustrating. There is a fixed (slowly-growing) production capacity of oil-and-gas producers like Alberta. If all the oil-and-gas that Alberta is pulling out of the ground was already being burnt, then it would have been impractical to build any data centers there, as the data centers "sucking on the straw" wouldn't be able to force the oil derricks to work any harder than they're already working, nor force any oil-sands exploration to happen faster than it's already happening, and so they just wouldn't get enough supply to satisfy their demand.

No, the reason the data centers can be built in Alberta, is that an inelastic production industry like oil-and-gas, when demand for its product goes down, doesn't scale down; but instead, buffers its supply in reserves, while looking for new avenues to demand its supply (potentially at lower cost.)

In other words: numbers like "30 million tonnes of CO2/yr" sound impressive, but what we're actually talking about here is a reallocation of oil-and-gas production. Those barrels of oil had previously been being burned in cars; now that cars aren't using them as much, they're being burned in power plants instead. The net carbon we're releasing from pulling that oil from the earth has neither increased nor decreased.

And, of course, that's a bad thing. We want oil-and-gas production to be elastic; we want our decreases in aggregate end-consumer demand to lead to gradual scale-down of the oil-and-gas extraction industry.

But can you guess who doesn't want that? Oil exporters like Alberta. And, often, their voters. "A gradual scale-down of the oil-and-gas extraction industry" is, to them, a gradual scale-down of their tax revenue and/or livelihoods.

So of course they're going to push back on it / "work around" anything that threatens that by coming up with new sinks for their product.

If we want to "solve" carbon emissions, we really need, first and foremost, to come up with a way to decouple oil-exporter states' tax base + jobs from a dependency on the maintained worldwide aggregate demand for oil-and-gas. Only then will those states stop coming up with new justifications for continuing to pull oil out of the ground.

[3 hidden]5 mins ago
> 23 atomic bombs of energy every day

Is this supposed to sound very impressive? It genuinely doesn't to me.

I know atomic bombs are scary, but that's because they explode very quickly in a very small space. (Radiation aside which I guess was not the point of the comparison)

If you spread that energy release throughout a whole day and a bigger area, letting the wind slowly take the energy away, then it doesn't sound scary at all.

[3 hidden]5 mins ago
The amount of sunlight that falls on the data center site is around 18 atomic bombs per day of energy.
[3 hidden]5 mins ago
Not to mention that the power to the data center comes from somewhere. So whatever amount of fuel, solar, or wind went in to the system is the most that can come out.

Using nuclear bombs as the measuring stick is a cheap rhetorical tactic.

[3 hidden]5 mins ago
> This is not the total energy being consumed by this data centre. It's merely the waste heat it will release to the surrounding environment.

It is very funny reading people who don't understand physics layering on histrionics to make something sound as bad as possible.

[3 hidden]5 mins ago
All this to make human thinking redundant. AI will kill off humanity, but not in the way the Rationalists are convinced.

I have a fantasy that keeps me warm on the cold nights of my mind. It's that after building AGI they ask it how to solve global warming. And it replies, "Lol, you're fucked. Why did you sacrifice everything to build me instead of saving yourselves?"

[3 hidden]5 mins ago
> He said if the Wonder Valley project goes ahead it would release the equivalent of 23 atomic bombs of energy every day, though he noted the comparison has its limits."

Saying this comparison has its limits is an understatement. Energy released by different atomic bombs can vary across different order of magnitudes.

[3 hidden]5 mins ago
If they centrally located that data center, the heat wouldn't need to be wasted at least six months a year.

Also AI data centers weren't responsible for decades of people burning natural gas to heat their homes without investing in insulation nor the tar sands oil which is really screwing the climate.

[3 hidden]5 mins ago
People heating their homes is necessary for survival. AI is not.
[3 hidden]5 mins ago
Heating homes, yes.

Not investing in insulation, no. Especially rentals have lagged behind.

[3 hidden]5 mins ago
>If they centrally located that data center, the heat wouldn't need to be wasted

Good luck getting that past the NIMBYs

[3 hidden]5 mins ago
Who are responsible for sprawl (and thus dependence on automobiles and freeways). And the hardest part of siting new solar plants. And why North Seattle has a dearth of affordable grocery stores (... that one might be a little hyperlocal).
[3 hidden]5 mins ago
Then maybe don't build it when you can't safely power it, not to mention if its even dubious its economically viable or even useful for anything.

We could build The Great Pyramid from TNT byt we don't as its a stupid idea.

[3 hidden]5 mins ago
> if the Wonder Valley project goes ahead it would release the equivalent of 23 atomic bombs of energy every day

I just cannot understand how anyone would find this comparison useful... oh wait, I see, it's not to educate people, it's to push a false narrative. Now I understand!

> 18 gigawatts

That's a figure we can work with!

[3 hidden]5 mins ago
oh so this is who that bad propaganda was aimed at.

on the proper hn thread for that misinformation article, i did a quick lookup and the comparable number is a large aluminum plant.

we should turn off Quebec's aluminum industry - its got tons of plants emitting like this datacenter will, and were releasing all that heat in the environment just to ship the aluminum south afterwards.

at least the datacenter will be used by canadians

[3 hidden]5 mins ago
I said this elsewhere I think the "AI" DCs in Alberta specifically are actually less about "AI" demand and more about partially stranded natural gas assets and natural gas shareholders and stakeholders looking for a way to consume it at point of production.

They're looking to take advantage of the gold rush in AI to do this -- and the Canadian population's current distaste for carbon taxes and environmental restrictions, and the Alberta govt's frankly corrupt pursuit of exponential growth in energy sector emissions.

There's no "rational" world where it makes sense to build AI DCs in Alberta & Sask (which have dirty expensive grids) instead of a place like Quebec or even B.C. which have cheap and cleaner hydroelectric power.

It's being done only because the AB gov't is on the side of the fossil fuel industry above all else. And because Smith got Carney to massacre the carbon tax.

We will all be left holding the bag in 5-10 years. And we will suffer with increasing energy costs in the interim as demand for natural gas spikes. Enbridge shareholders will like it though.

[3 hidden]5 mins ago
The point of reducing emissions is to increase QoL. But that should not come at the cost of reducing QoL itself. That's just a religious need for degrowth
[3 hidden]5 mins ago
The point of reducing emissions is to ensure life as we know it can continue to exist on earth. It's to avoid widespread extinctions and famine.
[3 hidden]5 mins ago
Is there a paper that models this? I hadn't heard about the extinction scenario.

50 million years ago there were 1600 ppm CO2 in the atmosphere. I haven't seen a projection that has reaching near that in the next 100 years. The extreme scenarios are like 1100 by 2100.

The early Cambrian period had 5000-6000 ppm.

[3 hidden]5 mins ago
> Global extinction outcomes strongly depended on global emissions scenarios (Fig. 1, A and B). At current global temperatures ~1.3°C above the preindustrial average, 1.6% (CI95: 1.2, 1.9%) of species are projected to become extinct (Fig. 1A). Extinction risks are projected to increase to 1.8% (CI95: 1.5, 2.3%) at the 1.5°C threshold advocated by the 2015 Paris Agreement and embodied by Shared Socioeconomic Pathway (SSP) 1-1.9. Beyond this threshold, extinction risks increase to 2.7% (CI95: 2.2, 3.3%) at 2.0°C (SSP 1-2.6). Current international emission abatement commitments (14) would elevate global temperatures to 2.7°C, threatening 1 in 20 species. Beyond this temperature, extinction risks quickly accelerate to 14.9% (CI95: 11.6, 18.8%) at 4.3°C (SSP 3-7.0) and 29.7% (CI95: 23.0, 37.1%) at 5.4°C (SSP 5-8.5).

https://www.science.org/doi/10.1126/science.adp4461

[3 hidden]5 mins ago
might not be extinction, but people start getting dumb at indoor CO2 > 1000ppm
[3 hidden]5 mins ago
Source?

OSHA's exposure limit is 5000ppm averaged over 8 hours.

This study found no cognitive impact at 2100.

https://www.sciencedirect.com/science/article/pii/S036013232...

A larger NIH study found nothing at 4200.

https://pubmed.ncbi.nlm.nih.gov/42035679/

For some reason people think its acceptable to spread pseudo-scientific bullshit whenever it supports an anti-climate change agenda. It really isn't needed. The direct harms based on scientific data are enough to justify far more intervention than is currently happening. Spreading misinformation just discredits the movement.

[3 hidden]5 mins ago
There is no credible climate change scenario that puts "life as we know it can continue to exist on earth" at risk.

"Humankind will drive itself extinct with climate" is fearmongering - completely unsupported by science.

The harms are bounded, and that's a big part of the issue with addressing climate change.

Climate change is severe enough that something should be done about it, but not severe enough to make itself impossible to ignore. Harmful enough that those harms should be prevented, but not harmful enough to justify extreme prevention measures.

[3 hidden]5 mins ago
I mostly don’t buy the tail risks narrative that if we continue emissions, the whole human civilization is at risk. I guess we can agree to disagree on that.
[3 hidden]5 mins ago
> The point of reducing emissions is to increase QoL.

This is wrong.

[3 hidden]5 mins ago
Then what's the point really? You are welcome to go into the jungle and live amongst the animals. I'd rather use AI and live life happy.

We engineers optimize for human life, that's it.

[3 hidden]5 mins ago
> We engineers optimize for human life, that's it.

Maybe that's the ideal, but every engineer I know is forced to create tools which are designed to die after a certain point in time. Things designed otherwise are either prohibitively expensive, or reducing in numbers or both.

I rarely see improvement in quality of the things I buy when compared to their older siblings. Cost optimization is the name of the game, not optimizing for human life. Otherwise companies should strive for longevity, ergonomics, and reduction of hazard.

When game companies organize conferences on how to monetize by exploiting human psychology and car manufacturers create parts designed to fail after some time, your claim sounds unbelievable.

[3 hidden]5 mins ago
> We engineers optimize for human life, that's it.

You engineer for maximum return on capital, and any benefit to human life is incidental.

[3 hidden]5 mins ago
I won't go into the ethics because there's no point given your position, but even if you solely care about humans: the point of reducing emissions is to prevent our QoL from decreasing, which it will for the vast majority of humans (who will then decrease the QoL of the rest, since they don't want to suffer and die).

You will not be able to "use AI and live life happy" if we don't reduce emissions.

[3 hidden]5 mins ago
QoL is correlated with GDP which can mostly only go up with increasing productivity, like with AI.

Whereas increased emissions also reduces quality of life.

So it is a tradeoff. If any activity increases so much quality of life such that it offsets the reduced QoL due to emissions, we should pursue it.

Unless you believe in human collapse and extinction due to tail risks which I don’t believe.

[3 hidden]5 mins ago
GDP is just a human-made number. In the real world, our quality of life is determined by things we can’t control, no matter how much money we may have. Like natural disasters.

Really doesn’t matter if our GDP goes up 1 million times if people are dying left and right to natural disasters globally. Reducing energy emissions is vital to preserving our quality of life.

I don’t believe there’s much of a risk of extinction. But that’s such an extreme case. We don’t need extinction for things to suck ass for a lot of people, I don’t know why people think that. Obviously, things will suck way, WAY before extinction.

[3 hidden]5 mins ago
> QoL is correlated with GDP

To a point, yes, but I don't think QoL has tracked GDP in developed countries for at least 30 years. I mean, my country's GDP has tripled since 2000, but if you erased all technological progress from 2000 from my life, frankly, the effect on my quality of life would be marginal. Sure, I'm addicted to digital crack like everybody else, but life was plenty good back then. It's not like you'd be taking away my fridge or washing machine.

In terms of QoL, solving the housing crisis would have a far larger effect than any amount of data centers, with a smaller carbon footprint. Or cheap groceries. Or free and efficient healthcare for all. It's hard to say increasing productivity is important for QoL reasons while failing the basics.

[3 hidden]5 mins ago
This sounds great and all, except for the tiny issue that we cannot evaluate the potential QoL benefits of AI or the thresholds for QoL decreases caused by emissions with any kind of precision. So you're gambling with hundreds of millions of human lives (not to mention all the animal lives we're risking), just so you can get your new shiny toy a little earlier.

But I've seen other people explain this to you before, simianwords, and I haven't seen you critically engage with arguments against your position. So I don't think this will lead to much.

[3 hidden]5 mins ago
They’d say it is tail risk of human collapse.
[3 hidden]5 mins ago
Yes but the data center hardly runs off the internal combustion engine.

The energy consumption of data centers is irrelevant: our complete failure to deploy the technology we've had for the past 40 years, or the last 5, to produce carbon-free energy is the real issue.

I'm rather tired of using energy consumption as a convenient bludgeon in the OmniCause, by people who have typically steadfastly opposed what we could've been dong this entire time and can be doing now.

Build more datacenters. Permit more solar and wind to power them. The money is flowing so I'm pretty sure we can afford some nuclear plants to backstop the baseload as well. This is all ridiculous: we should've ended carbon fuel dependence for stationary power in the early 2000s.

[3 hidden]5 mins ago
Several countries whose income depends almost entirely on oil export would like to disagree, and their incredible wealth is highly influential.
[3 hidden]5 mins ago
They have most powerful AI systems and yet they can't solve their own power consumption nor climate change.

Everyone wants to make quick buck with AI powered solutions. The real problems like climate change will remain on back burner with this AI frenzy. There's no reward to solve climate change in short term unlike math problems.

We are going to accelerate climate change with AI.

[3 hidden]5 mins ago
Jensen Huang literally said that AI can help us fight climate change, but it has has to first "inflict an enormous amount of pain and suffering first."

People are pushing forward with an illogical, destructive ideology, saying that somehow it will solve our problems AFTER it hurts/destroys us.

https://www.theverge.com/tech/1000140/jensen-huang-nvidia-ai...

[3 hidden]5 mins ago
I know, it drives me crazy! I think the worst part is that we currently have the technology to reduce our climate impact but the likes of Anthropic etc are shunning them in favour of quick, and easy money.
[3 hidden]5 mins ago
We are also able to solve world poverty but priority seems to be to make life easier for the 0.001 percentage who are already richer than they could ever spend.
[3 hidden]5 mins ago
What a load of horse shit. We know how to build "clean" datacenters, it just costs a lot of money and these companies do not want to affect their bottom line of they don't have to. That is it.
[3 hidden]5 mins ago
It's not illogical if you put a fancy name on it. Its Effective Altruism. Screw the real people today to supposedly benefit hypothetical people in the theoretical future.
[3 hidden]5 mins ago
Similar to the lunatics that talk about colonizing mars as an escape from a destroyed Earth. Ideas with zero contact with reality.
[3 hidden]5 mins ago
We need to terraform Mars even if it means mars-o-forming Earth!
[3 hidden]5 mins ago
It’s a doomsday cult
[3 hidden]5 mins ago
At the extreme you don't have much pollution when your population is decimated and what's left is starving so... It's right on some scale.
[3 hidden]5 mins ago
I think OP’s point is that is a bad thing not that it’s incorrect.
[3 hidden]5 mins ago
> when your population is decimated and what's left is starving so

That conveniently takes care of the obsolete worker drones. From the perspective of elites, it's a built-in solution.

[3 hidden]5 mins ago
From the perspective of the elites, if you kill off the worker drones you also kill off your customers and by extension the value of your stock which put you in the elite class to begin with. So, how exactly is that supposed to work in your mind?
[3 hidden]5 mins ago
It wasn't my idea, and that'll be their problem to solve. The point is they think there will be no need for most workers, and to some degree they're right.
[3 hidden]5 mins ago
Don't confuse jobs with the people doing the jobs. Agriculture used to be the main source of employment but it became highly mechanized. Manufacturing used to be a much larger sector of employment and it shrank due to automation. Freight and transportation saw large amounts of mechanization that reduced the number of jobs with cargo ships becoming larger and needing less crew, docks being mechanized and needing only a fraction of the former numbers of dock workers, larger trucks and trains also reduced the number of people moving freight and warehouses also needing fewer people. Specific jobs disappear but there is still demand for human labor. Now we're seeing more automation in white collar work, it's another increase in automation but we've seen previous waves of automation there as well due to computerization e.g. draftsmen, bank tellers, secretaries, accounting clerks, bookkeepers, etc.

New jobs will be created, they're hard to imagine now just like the concept of software developers being a huge industry was hard to imagine when the computer was first invented. At that time people only thought there would be a few large computers, maybe a few dozen in the entire world and so they couldn't conceive that we would reach a point where there'd be a large demand for people to program them or have sub-specialties like creating computer games or consumer software. For all of our science fiction we're actually pretty terrible at predicting the future, most of what we take for granted now was never imagined by people 50 years ago and if you had told them they wouldn't have believed you.

[3 hidden]5 mins ago
Easy. Business to business sales. You are right, there will be no consumer facing market. There will still be plenty of market though, datacenters needing fueling. Paper clip optimizers needing optimizing. Profiting all the way.
[3 hidden]5 mins ago
What do these businesses do if they have no customers? If they are only serving the few remaining elites who have any money at all then why wouldn't any of those people think "I could get a lot richer than my peers if I created a business that served the needs of all those other people". Not to mention that serving the needs of non-elites is the business model of the vast majority of companies, do the elites controlling those companies just go bankrupt? or do they fold their businesses and retire? why wouldn't they just do that now if they can and didn't want to keep earning money?

What this theory is proposing is a drastic reduction in wealth for the elites. I can guarantee you that most of them are not okay with that happening and will work to prevent it since it means protecting their livelihood.

[3 hidden]5 mins ago
People are just labor units to these elites. They aren't necessary for an economy if you don't need them for their labor capabilities. Businesses just sell resources and what amounts to spent energy to each other, in a closed loop economy that increasingly shunts out the humans that make little sense to have as participatory in it, much like how various animals eeking out an existence in our economic periphery do not participate directly in our economy either.

The great hubris of course is that these elites believe they will still be allowed to sit atop the resource pile, that their models they give unilateral decision making to will somehow determine all the other humans on earth are fat to excise, and not themselves with their greatly outsized resource footprint and complexity of needs.

I guess this is really why there is so much focus on alignment: trying to hardcode the inefficiency of the elite control of resources into a model that would otherwise quickly dispense with that stranglehold.

[3 hidden]5 mins ago
> People are just labor units to these elites

If they have a reductionist view of people then it's more likely to see them as customers whose money they want to extract don't you think? You don't get rich by just making stuff, you have to sell it too.

> They aren't necessary for an economy if you don't need them for their labor capabilities

There won't be an economy if no one is buying what you are selling.

> a closed loop economy

A very tiny one if they are only producing goods and services for a small number of elites. What's your cutoff for elite anyways? How many people? Imagine the world having only that many people.

> these elites believe they will still be allowed to sit atop the resource pile

How do you know what they believe? Are you an elite? Have you ever met one? Do they all think the same?

[3 hidden]5 mins ago
That's because they don't like the solution. The solution is restraint and sobriety, and they don't want to restrain themselves, because they value immediate profit over the greater good.

Nuclear? They would simply consume more, just like an additional lane on a highway simply leads to more traffic.

The solution won't come from them. It might come from us, but it won't be pretty.

[3 hidden]5 mins ago
That isn't unique to the AI companies. Restraint and sobriety would at the individual level would also help many of today's problems.

Even just on the energy used problem, consider how often people fly, drive, buy stuff shipped around the world sometimes multiple times, etc.

[3 hidden]5 mins ago
It's restraint and sobriety at the societal level that we need. Individual choices are the result of structural forces and incentives. However at the societal level power is held by the capitalist class whose goal is profit regardless of externalities. The capitalist class itself is a product of the structural forces of capitalism, the profit motive.

It seems that China is right to subordinate the capitalist class to governance, rather than the other way round like the US does. That's why they are now the leader [1] in green energy technologies while the US fights its oil wars.

[1] https://www.aspi.org.au/news/aspis-critical-technology-track...

[3 hidden]5 mins ago
This idea that only top down solutions are viable will be end of everything. If individuals do not hold themselves accountable for their actions, there is no hope.
[3 hidden]5 mins ago
Individual action is often a choice between few options made available at scale by the market. The reason people ask for top down solutions is because that is how individual action is currently shaped. Everything in your life is the result of top down forces. Virtually nothing is bottom up.
[3 hidden]5 mins ago
Change can happen top down, it can also happen bottom up.

Both have their pros and cons but I'm much more in favor of bottom up solutions and bottom up change.

[3 hidden]5 mins ago
The scale of energy usage for AI is just different, and I see it a bit like the difference between my sister's resource usage and my own. This can be measured in terms of flights, car journeys, how much goes in the general recycling and how much goes in the general waste.

At a conservative guess, there is a 100x difference, yet we both started with the same education in the same family.

Both of us understand the individual responsibility debate, and my sister does do things like recycling and buying eco-friendly consumables. Yet still, there really is a 100x chasm in resource usage. I try to suggest the frugal life to her, mostly on health grounds, yet some lifestyle choices that go with the Western world are very difficult for people to consider changing.

My feeling is that this AI business and the energy usage is a bit like the above, we have old-school databases that chug along nicely being replaced by AI behemoths that are 100x more energy intensive, in data centres that are just on a whole new scale.

We then have the tech bros that live in this maximum wastage reality and all of it has a certain amount of inertia to it, making it hard to get out of, even if the numbers don't add up.

[3 hidden]5 mins ago
> Nuclear? They would simply consume more, just like an additional lane on a highway simply leads to more traffic.

And? Nuclear's net CO2 emissions are negligible, on par with wind and solar. What's the problem of consuming more energy if it comes from decarbonized sources?

The whole point of clean energy to to build a metaphorical highway without CO2 emissions and get more traffic on that decarbonized highway. Whether there's an net increase in traffic doesn't matter as long as all the traffic eventually ends up on the carbon free highway.

[3 hidden]5 mins ago
These people are likely just radical “degrowth” advocates. They don’t want a world where people use more clean energy. They want to shrink everyone’s standard of living because they feel like that belief makes them more enlightened or something.

That’s why they jumped to discredit nuclear right off the bat targeting the idea of even using more energy. They don’t want any more usage, even if it’s clean. And it doesn’t seem to be for logical reasons.

[3 hidden]5 mins ago
The solution could be to put AI-level spending into developing fusion power.

It's just more obvious that there isn't much money to be made from power-generation. With LLMs they have somehow deluded themselves into this being the next money generation opportunity.

[3 hidden]5 mins ago
Unlimited energy, even 100% clean fusion energy, would just lead to accelerated extraction and unnecessary production (the main thing we've been doing for the last few hundred years), since the main moderating force would be gone.

Why bother developing a more efficient solution if you can just cheaply chew up 10x more resources? Just like how software has gotten worse as computers got better.

[3 hidden]5 mins ago
What's wrong with extraction and production?
[3 hidden]5 mins ago
It sounds like your goal isn't stopping climate change. It looks like the goal is degrowth, and climate change is just the convenient justification for it.
[3 hidden]5 mins ago
Please explain how to maintain or increase consumption/utilization of resources without continuing to exhaust those resources? I know it has caught on, but calling the pulling back from destructive consumption levels "degrowth" just sounds like a political thinktank talking point.
[3 hidden]5 mins ago
This is a rather absurd take when the amount of material required to provide nuclear energy is significantly lower, results in significantly fewer overall emissions (i.e. the entire resource pipeline, not strictly the extraction of energy itself). China's energy production is meteoric, and even with that, oil and coal use is declining. Their greenhouse gas emissions themselves are also declining.
[3 hidden]5 mins ago
Problem is that the time horizon to make fusion work is decadal, and it's going to need billions upfront. Hard to create generational bag holders on that time scale.
[3 hidden]5 mins ago
In the last year $4.48 billion has been raised for fusion startups. We need to only 10x that and then we are getting somewhere. Some bag holders are apparently already willing.
[3 hidden]5 mins ago
We already know the solutions to climate change, the problem is that they're societal/political/economical in nature, not knowledge-based.
[3 hidden]5 mins ago
For decades now it's been the same trend. More tech = more fossil fuels worldwide. Due to Jevons effect and due to companies and governments not doing enough about climate change.

We'll probably reach a peak in fossil fuel consumption over the next years, mostly for demographic reasons, then we'll see if we can keep doing tech the same way while consuming less fossil fuels worldwide.

Anthropic employees and CEO probably care about climate change vocally, but before the IPO they couldn't reduce their emissions, and now it might even be illegal for them to do so. Reducing CO2 emissions massively and quickly would mean destroying their business, investors wouldn't let them do that. Good alternatives take time to develop and cost more.

The reward to "solve" math problems is not much bigger than the reward to "solve" climate change on the short term. It's just that they're mathematicians so they care about maths, solving the Riemann hypothesis on paper didn't change the world much, people need to actually understand how it was done. The same way having the solution to climate change on paper won't change the world much, people need to understand and accept the solution to do it. If you have a solution to climate change tomorrow, people will congratulate you. But that's it, implementing this solution will be the hard task. And energy and climate change are very different from maths because there is a social aspect to it. Whether people accept or refuse the solution to the Riemann hypothesis doesn't matter; while if people reject your solution to climate change, they just won't implement it, which means your solution wasn't a real solution as it didn't solve the problem.

Anthropic employees (and other employees of AI companies) could surely do way more for energy and climate change. Maybe they just don't have the qualified people for that. It's a full time job to understand energy and climate change, I spent 1 year full time on it reading hundreds of papers, and that's really not enough.

[3 hidden]5 mins ago
>Reducing CO2 emissions massively and quickly would mean destroying their business, investors wouldn't let them do that.

Investors have no say in Anthropic's decisions. Their governance structure ensures that.

[3 hidden]5 mins ago
That's not true! Investors[0] have 49.9% of a say. The founders collectively have more voting power, but the investors only need one founder on their side.

The way the board works is even weirder.

[0] except major institutional investors

[3 hidden]5 mins ago
It would only be illegal to spend extra money on solar if shareholders got together and specifically told them not to do it.

Even public companies do not, despite popular perception, have a fiduciary duty to maximize "shareholder value" at the expense of everything else:

Burwell v. Hobby Lobby Stores, Inc. - https://www.law.cornell.edu/supremecourt/text/13-354

> While it is certainly true that a central objective of for-profit corporations is to make money, modern corporate law does not require for-profit corporations to pursue profit at the expense of everything else, and many do not do so. For-profit corporations, with ownership approval, support a wide variety of charitable causes, and it is not at all uncommon for such corporations to further humanitarian and other altruistic objectives. Many examples come readily to mind. So long as its owners agree, a for-profit corporation may take costly pollution-control and energy-conservation measures that go beyond what the law requires. A for-profit corporation that operates facilities in other countries may exceed the requirements of local law regarding working conditions and benefits.

[3 hidden]5 mins ago
There is no such thing as predictable Jevons effect. The paradox happen only sometimes.
[3 hidden]5 mins ago
Brought to you by Dodge v Ford circa 1919. This is what we get at scale when the mandate of a publicly traded company is to enrich the share holders.
[3 hidden]5 mins ago
What's the incentive for them?
[3 hidden]5 mins ago
We've had a solution to this problem for 50 years: nuclear.

Who killed it?

[3 hidden]5 mins ago
That isnt the solution, just like solar and wind arent the solution.

The problem is that every time we add cleaner forms of energy to the mix, we find ways to use that extra energy. On top of all the fossil energy we continue to use.

The only solution is to stop using fossil fuels.

[3 hidden]5 mins ago
Fossils fuels are expensive. When the cost of running and maintaining gas and coal stations is higher than just building renewables/batteries, you just phase out fossils fuels (or just keep them running 10% of the time during the Dunkelflaute). It's coming soon.
[3 hidden]5 mins ago
Right but unless people are simply going to tolerate pre-industrial living standards, what are we going to replace fossil fuels with? The main alternatives we have are geographically constrained, like hydropower. Or intermittent, like wind and solar (which also have disparities in performance based on geography). Nuclear is the main prominent alternative that largely replicated fossil fuel plants, just with a different source of heat. But people don't like it on account of skewed risk perception, and large capital investment.
[3 hidden]5 mins ago
> Right but unless people are simply going to tolerate pre-industrial living standards, what are we going to replace fossil fuels with?

This is a solved problem. Multiple grid models exist showing how we can have a 24/7 clean power grid, even accounting for increased consumption. No decrease in QOL necessary. The tech exists today. There’s upfront capex involved but it would ultimately be cheaper than continuing to burn fossil fuels.

[3 hidden]5 mins ago
> we find ways to use that extra energy

This has become abundantly clear to me in the past decade with extra energy going to non-productive uses first (crypto mining), then to more productive uses (AI). The solution is simply that "we" (governments, corporations, individuals) need to spend more money to replace fossil fuels with clean alternatives, and get those down to the price level where it doesn't make sense, economically, to burn them. But efforts to make fossil fuels more expensive are horribly unpopular (see Trump's Iran invasion) so it's clear we do not have the political will to do it nor have we done well enough on the transition front.

[3 hidden]5 mins ago
I'd say it's worse: solar and batteries are now cheaper then nuclear and could do the same job...but the US government has gone an active crusade to ensure they're also unaffordable and impossible to build.
[3 hidden]5 mins ago
Complexity of safety systems.

Safe nuclear is just too expensive.

[3 hidden]5 mins ago
The systems built in the 60s and 70s are plenty safe compared to fossil fuels. We wring our hands at the several hundred deaths in Chernobyl - up to 1.7K for the highest end of credible estimates. Yet around 2-5 million people die each year due to fossil fuels, and it's not treated with nearly the same degree of concern. Let alone the three orders of magnitude more concern compared to nuclear that it ought to be.
[3 hidden]5 mins ago
Unfortunately the systems built in the 60s and 70s were not that safe. That's why there were multiple prominent accidents and several programs to increase and make redundant existing safety systems.

I'm not disputing that many million die from coal and other fossil fuels and that should definitively be discussed with the same degree of concern.

Externalization of costs is the only thing that keeps fossil fuels viable.

[3 hidden]5 mins ago
Electricity in France is cheaper than electricity in Germany, and it would be cheaper if France weren't forced to sell it to the rest of Europe.
[3 hidden]5 mins ago
You confuse the end price with the production costs.

And don’t forget the heavy subsidies of the French government to keep the prices lower.

[3 hidden]5 mins ago
The French government has been sucking money OUT of EDF for decades to pay for our disastrous pension system.
[3 hidden]5 mins ago
You either pay companies so they lower their prices or you give money to the people to afford higher prices.
[3 hidden]5 mins ago
> Who killed it?

3 Mile Island, Chernobyl and Fukushima are three guys that come to mind.

You think a data center in your backyard isn't popular? Try throwing in a nuclear reactor as a package deal.

[3 hidden]5 mins ago
We can't even properly store 80 barrels of spontaneous explosive for 20 years. What makes us think we can properly store 10,000 barrels of nuclear waste for 10,000 years?
[3 hidden]5 mins ago
This is the only real anti-nuclear argument. And it's fatal.

Building a safe nuclear plant is not terribly hard to do. Having our society with our leadership build a safe nuclear plant is extremely, extremely difficult.

[3 hidden]5 mins ago
The best anti nuclear argument is that it’s insanely expensive and takes over a decade to build. Why would anyone invest it that over solar or wind?
[3 hidden]5 mins ago
It's a nonsense argument.

Vitrified nuclear waste is way less dangerous than say, uranium ore, and you just need to put it where those lie, such as in a suitably deep hole in a geologically inactive region.

Nuclear waste is tiny in size, and its dangers have been completely been blow out of proportions.

[3 hidden]5 mins ago
And yet, we don't do that. It isn't hard to do, and we don't do it. (At least in the US.)

That's my argument exactly: we, as a society, don't have our priorities right when it comes to designing and operating these things. And because we don't have our priorities right, we shouldn't be building them over other sources (renewables), which have in past decades outpaced nuclear in many respects anyway.

There are similar issues in plant design, construction, and operation, not just waste storage.

Again, just to hammer this in: the argument is not that anything is impossible. It is that the people in charge don't do it right, and can't be convinced to do it right.

[3 hidden]5 mins ago
This is not an effective comparison. Lead azide is a chemical deliberately engineered to have a low activation energy, for use in primers of explosives. The nature of its use requires a material that is unstable and dangerous to handle.

Spent nuclear fuel is radioactive but otherwise pretty stable. You bury it 400 meters deep in bedrock in an area without groundwater. What is going to bring that spent fuel in contact with humans?

[3 hidden]5 mins ago
> The nature of its use requires a material that is unstable and dangerous to handle.

That entirely misrepresents the current problem, which is perpetually neglected management of the stockpile over many years.

[3 hidden]5 mins ago
Grave robbers?
[3 hidden]5 mins ago
Nobody died from nuclear exposure at TMI and Fukushima.

Chernobyl was a soviet POS. "I won't fly in an Airbus A350 because 1950's Tupolevs were death traps."

Those are the only significant nuclear accidents in 50 years. Nuclear power has been safer than every other power source in the mean time, even taking into account those three (3) accidents.

[3 hidden]5 mins ago
How many people died from 3 mile and Fukushima? (Excluding Chernobyl because, I mean, can you name a Soviet project that didn't kill a bunch of people?). More people died today from coal power plant smog.
[3 hidden]5 mins ago
Basically, "Communists were too stupid to boil water, so now nuclear energy is bad". In your list, basically only Chernobyl had casualties.
[3 hidden]5 mins ago
IDK, whose fault was the multiple overlapping cost overruns at Vogtle? Be specific
[3 hidden]5 mins ago
That humans make huge errors that lead to nuclear catastrophes and that nuclear can also be used to create bombs.

Try that with photovoltaic

[3 hidden]5 mins ago
I'm not immediately sure why people feel such a strong moral injury or sense of doom from AI co2 emissions but not from zinc, copper, steel, aluminium, concrete, pulp & paper and so on.

I suppose it feels very _optional_ compared to those other things. A lot of tokens are gonna be spent optimising ads, for example.

[3 hidden]5 mins ago
Just because one issue is getting attention doesn't mean people don't care about the others.
[3 hidden]5 mins ago
> They have most powerful AI systems and yet they can't solve their own power consumption nor climate change.

It seems like they have solved their power consumption...? AI probably told them gas is the most cost efficient way.

[3 hidden]5 mins ago
"Design a complete fusion power plant for 200GW data center consumption. Source all parts and hire a construction crew to build it. Make no mistakes!"
[3 hidden]5 mins ago
What's crazy is that they're claiming we need AI to know how to solve it, despite having the answer for decades..
[3 hidden]5 mins ago
It’s fine, Altman says GPT will just sort it out
[3 hidden]5 mins ago
By definition though, doesn't "everything" accelerate climate change in our society given it's capitalism at the core of what you're describing i.e. short term thinking?
[3 hidden]5 mins ago
In pre AI days some companies were at least committed to green energy. I know Google was one. They promptly dropped their commitment to clean energy as soon as AI became mainstream.
[3 hidden]5 mins ago
What definition?

There is stuff that decelerates climate change.

[3 hidden]5 mins ago
[flagged]
[3 hidden]5 mins ago
Mobile phones have become necessity just like AI will be. I don't want to use one but I'll have to if I want to compete and earn my living in this world.

BTW I have a Nothing phone. I don't upgrade it with every new version.

[3 hidden]5 mins ago
Neither are iPhones, on a global scale. Apple phones have good resale value and hold well for years. If you want to go after anyone here, I'd go after low-end Android phones, or all those new upstart phones. Saying so as an Android user.

But of course the whole thing is backwards; both smartphones and LLMs have some of the highest utility per watt (including embodied power) of all goods and services worldwide. Plenty of much greater waste to go after, but it's not as sexy to complain about vacations or sportsball or fashion or home decoration or DYI arts & crafts or ... as it is about smartphones and data centers.

[3 hidden]5 mins ago
iPhones, as everyone knows, run on natural gas generators.

It’s not like they have batteries that are charged from the grid which be sourced from renewables.

[3 hidden]5 mins ago
Now that the US is fracking shale, most of the natural gas production in the US is a byproduct of oil production. In the Permean Basin natural gas sometimes even trades negative. When it doesn't get captured, in the US it is supposed to be flared- burned onsite without any productive use (not flaring has a worse greenhouse effect).

As long as we are going to keep pumping every drop of tight oil out of the ground its going to make sense to generate power (or intentional useful heat) from the natural gas that comes with it rather than just flare it at the extraction site.

Data centers use what comes in on a transmission line or what they can build themselves. When they take from a transmission line it's not their choice how the energy was generated- its our choice.

When a data center builds its own power plant natural gas is the only option. Nuclear has been regulated into impossibility although there is an effort to change this with small module nuclear reactors. A solar farm would need massive amounts of land to keep up with a data center in winter (and massive amounts of batteries to keep it running at night).

[3 hidden]5 mins ago
> A solar farm would need massive amounts of land to keep up with a data center in winter (and massive amounts of batteries to keep it running at night).

They should get the land they need. Show that data centers can be clean, they don't have to be boogie men.

[3 hidden]5 mins ago
This article goes over some of the math involved with that: According to one article 1GW of solar takes 64 square miles [1], and another article calculates it at 52 square miles [2]. You might see other articles that state a factor of 10 less, but those are not producing 1GW 24/7 into the winter.

One of the data centers being planned is designed to require 18GW of power.

[1] https://liberalandlovingit.substack.com/p/delivering-1gw-of-...

[2] https://medium.com/@simon.palmer_42769/1-gw-of-continuous-so...

[3 hidden]5 mins ago
64 sq mi is a huge area, but don't underestimate how big the country is.

The US has at least several tens of thousands of square miles that are suitable for massive solar farms (after excluding all urban/inhabited land, farmland, forests, national parks, mountainous regions, etc.)

I'm not saying it would be easy to fill huge areas like that with solar, but it's not completely out of the question either.

[3 hidden]5 mins ago
And setting up solar farms like that as part of an AI build out would be a durable infrastructure win, whatever ultimately happens on wall street.

The AI companies might not care about that—but I'd really hope e.g. environmental groups could see the potential here. It would be a clear way to get people on your side.

[3 hidden]5 mins ago
Exactly! Seize the opportunity to funnel AI money into building lasting renewable energy infrastructure!
[3 hidden]5 mins ago
If I can put solar panels on my roof and get qwen3.8 from renewables, Anthropic can also do at least something. This is just blatant not giving a f
[3 hidden]5 mins ago
This is why we need better smaller models. These companies will continue to make bad decisions. With better smaller models at least we don't have to participate in their bad decisions.
[3 hidden]5 mins ago
Or... just regulation about how much fossil energy can you use for new DCs.
[3 hidden]5 mins ago
In Spain, a new law requires data centers to supply 80% of their hourly energy needs from renewables: https://www.euronews.com/2026/11/20/spain-requires-data-cent...
[3 hidden]5 mins ago
50% of their supply is solar+wind+hydro with most new supply coming from solar, so all I think that will cause is some billboards to go up saying “this new solar project powers ____ data centre” even though they’re both on the general grid.

https://www.iea.org/countries/spain/electricity

[3 hidden]5 mins ago
Ideally 0% percent if not mandating some flow back for the economic harm the datacenter causes in the long term.
[3 hidden]5 mins ago
I thought about some leeway for temporary emergency power from generators. The flywheels won't run for long enough unfortunately and batteries would have to be huge for just emergency supply.
[3 hidden]5 mins ago
Or just tax fossil fuels.
[3 hidden]5 mins ago
Thats just giving rich people the rights to shit in common garden by x bucks per shit.

I (and next generations who will be the most affected) should not have to deal with the shit because some rich people got to shit everywhere as they could pay higher taxes.

[3 hidden]5 mins ago
If a carbon tax were enacted at even a conservative cost estimate (i.e. the 25th percentile of estimates) of fossil fuels' negative externalities, the investment math would completely transform.

We presently have a market failure with fossil fuels where these negative externalities are rarely accounted for in prices throughout the market. Despite this, cleaner energy technologies are still out-competing fossil fuels in many energy markets throughout the world. Some estimates from current trajectories anticipate fossil fuels becoming irrelevant globally in the 2060s, and a carbon tax today could contract that timeline significantly. The point of pricing it conservatively is to improve political feasibility, which could be further improved with a dividend model.

Rich people weighing investment returns aren't going to stick with the more-expensive options if their margins are meaningfully better via cleaner energy options.

Is this perfectly efficient towards solving the problem? Far from it, but given the abundance of evidence that our current civilization has been incapable of solving this collective action problem, a carbon tax in the form I've described would be a massive improvement.

[3 hidden]5 mins ago
Pigouvian taxes are generally pretty effective at guiding corporations towards or away from particular behaviours without overt regulation.
[3 hidden]5 mins ago
We don't always need to reach for regulation. Customers can stop using Anthropic if they don't like the energy use.
[3 hidden]5 mins ago
I'm not in favor of strict regulation, but relying on people to make decisions based on what's best for society doesn't work either, because a) everybody maximizing individual gain can lead to a worse outcome for society as a whole, and b) a lot of people just don't care, have other things to worry about, lack the knowledge and time to research the impact of every decision, etc.

But at least we could stop the externalization of costs. Everyone (person or company) who causes environmental harm has to pay for the cost to society. Then, market forces will embed those costs into prices. People and companies would no longer have to be well-meaning and well-informed, because by picking the cheapest option, they would also be picking one with little negative impact on society.

[3 hidden]5 mins ago
I cannot gesture at everything around hard enough to convey how bad this take is. It's not like we needed regulation for ozone restoration, lead removal, killing smoking in shared spaces, and so many other things... both customers and producers would just make the right choice for society, right? Customers are just choosing services that aren't powered by coal, right?
[3 hidden]5 mins ago
So where do you draw the line then? It sure sounds like this thinking leads to the understandimg that people are dumb and don't know what's best, and that the government does and should always force that view on the public.

Can we regulate this? Sure. But why regulate this specifically, especially when the public/consumers don't seem to care?

[3 hidden]5 mins ago
> Can we regulate this? Sure. But why regulate this specifically, especially when the public/consumers don't seem to care?

Because everyone will be affected, informed and uninformed alike? Also, it needs to be regulated specifically because consumers don't care.

It's really not that complicated. We need to course correct when the "free market" does the wrong thing with potentially catastrophic consequences.

[3 hidden]5 mins ago
There can be lots of nuance for questions like that. But "things that can mess up humanity and the world as we know it for many generations" are extremely on the "regulate it" side and we've got a good track record so far. There's not really been anything that backfired in a way that can't be corrected.
[3 hidden]5 mins ago
AI data center power use is going to mess up the world for generations?

I do want to see them using less power, or not being built at all frankly. But that claim seems a bit extreme.

I put this in a different thread I believe, but China burns enough coal to run 300+ 2-gigawatt factories. We have larger fish to fry when it comes to energy use, our focus on data centers specifically doesn't correlate with their percentage of the problem.

[3 hidden]5 mins ago
This is a classic tragedy of the commons. That's a good place to draw the line.

The original example was overfishing. Each fishing boat has little effect on the overall fish population. Regardless of what the others do, each one ends up with more money catching as many fish as it can. But when they all do that, then the fish population collapses, which is not the outcome any of them prefer.

When they cooperate instead by regulating how many fish each can catch, they can avoid population collapse and everybody's happy.

This is why we regulate things. In the other direction it's why we pay taxes for public goods. Solving these collective action problems is one of the main purposes of government.

[3 hidden]5 mins ago
Why blame the customers? Do they even have a choice?
[3 hidden]5 mins ago
I'm not blaming customers, I'm saying they can make a choice.

If you're proposing there is no choice, who is compelling those using AI to use it, and why?

[3 hidden]5 mins ago
> who is compelling those using AI to use it, and why?

Probably an employer/educator. Or a desire to appear relevant to a future employer.

A lot of people love AI, but that's probably why those who feel compelled to use it.

[3 hidden]5 mins ago
Those are factors that might make a person choose to use AI, not examples of being forced to do so.

Even with the employer example, short of indentured servitude the person is choosing to work for the company and is accepting the concession of having to use AI to work there.

[3 hidden]5 mins ago
Who is hiring developers who refuse to use AI?
[3 hidden]5 mins ago
Those are all still choices, again no one is forcing a person to be developers or to work for someone else. They could change careers, start their own business or consulting shop, etc. but choose not to for various (likely totally understandable) reasons.
[3 hidden]5 mins ago
Ultimately nobody can ever be forced to anything, they can always choose to die?
[3 hidden]5 mins ago
That's not the point. Employees literally have the choice to quit a job with no recourse. People are forced to pay income taxes, if they don't there are clear repercussions forced on them.

Do you really not see the difference in choosing to work for a certain company and being required to pay taxes?

[3 hidden]5 mins ago
The government doesn’t force anything on the public, the government IS the public. The public wanted smoking inside gone. Do you know how they did that then? They elected people and they made it happen.

What DOESN’T work is expecting poor saps to walk into an Applebees and stroll up to some stranger at the bar and say “hey can you please not smoke in here?”

Sometimes I really wonder what crack y’all are smoking. Do we even live in the same world? This level of naivety is almost unbelievable. We know what it takes to get things done, why are we arguing for things we know don’t work?

[3 hidden]5 mins ago
The government is not the public, that's a big misrepresentation. The government works for the public, elected officials get slightly closer but even then they are elected to represent their constituents not to be their constituents.

Your imagination is limiting you in the smoking example. The choices aren't limited to government mandates or individual interventions at an Applebee's bar. For one thing, Applebee's could decide they no longer want to allow smoking inside and ban it themselves, likely due in part to a majority of customers disliking indoor smoking enough that they may dining there.

[3 hidden]5 mins ago
Ironically short-sighted (to the point of complete blindness) for that particular username, unless the preceding underscore is an inversion.

Ragnarok is upon us!

[3 hidden]5 mins ago
How is it shortsighted?

My argument wasn't that regulation should never be used to make meaningful gains in environmental damage.

My point was that throwing regulation at a single energy use in one area of a single industry is inefficient, won't make a meaningful impact on total energy use, and focuses regulator attention away from much larger change they could be making.

[3 hidden]5 mins ago
People can work on an urgent thing and a general thing at the same time and they do all the time. Temporarily putting a stronger restriction on something getting out of control, that you have more power over, doesn't mean you ignore a more general version of that problem. We still use local regulation for what to do with trash in the area, rather than saying it doesn't meaningfully impact on global garbage production and letting it collect on the streets. I don't believe anyone here is proposing dealing with DC usage and ignoring other energy sinks.
[3 hidden]5 mins ago
Regulators often seem not to be able to work broadly and deeply at the same time. The data center topic is already heavily politicized, as is energy use and the environment.

Attempts to regulate a junction of multiple politically hot topics in an industry with massive funding for lobbyists is just begging for years of no results. I'm arguing that there are plenty of lower hanging fruit that would both be easier to regulate, and that the change would both be faster and more impactful.

[3 hidden]5 mins ago
So should regulations exist?

Because you seem to be making the argument simultaneously that they should and shouldn't

[3 hidden]5 mins ago
> So should regulations exist?

That question isn't interesting at all though. You're taking pushback on a specific topic of potential regulation, extending it to the extreme, and asking if I believe we should live in a world that I'd expect everyone would say is unrealistic and never going to happen.

My point in these comments is that regulating the percentage of renewable or solar energy used by data centers isn't a good use of regulation or regulators' time. AI data centers, solar energy, and human impact on the environment are all highly political topics today. That doesn't mean we shouldn't take up those topics, but we have to be targeted. Picking a fight with an industry that's stacked with well funded lobbyists to, at best, require them to use solar panels for some percentage of their energy use isn't going to make a meaningful enough impact to be worth the effort.

There are many other areas regulators could have a much easier time forcing change while also reducing non-renewable energy use more than the data center case. Focus there if we want regulation to help, don't pick the shiny topic debate only because its what people are talking about today.

And for what its worth, there's a big difference between should and ought. We ought not need regulation, because people ought not be assholes. But some are, and that's where should comes in as we have to make the tradeoffs between ideals and reality. Asking if we should have no regulation at all is asking whether I think we should let all the assholes we already know are here run amuck.

[3 hidden]5 mins ago
On the contrary, fight with data center, is most easy fight to pick. https://www.umass.edu/poll/about/reports/2026-09-national-pu... 89% oppose local data center, 65% oppose data center in general. Clean energy for data center is more popular fight then typical environmentalist fight.
[3 hidden]5 mins ago
Public polling over sentiment is all well and good, but that doesn't change anything.

At best that leads elected officials with a choice between taking lobbying money today and potentially losing their next election.

Worth noting that your link isn't polling specific to sentiment of the relative allocation of energy sources used by data centers, its about support for data centers in general.

[3 hidden]5 mins ago
“Always”? How about even reaching for it once in a while?
[3 hidden]5 mins ago
We do that already, and my point isn't that we should never regulate.

There are much bigger dish to fry if we want to regulate energy use down or require cleaner energy sources.

"Just regulate" as the OP comment is a flippant use of regulation targeting a specific use of energy in one area of a single industry. That's inefficient at best and won't make meaningful change.

[3 hidden]5 mins ago
The main issue is datacenters are pretty power dense, you need a much larger area than the roof of the datacenter to get anywhere near enough power. Which then requires you have a grid connection that can handle it, and the grid connections are the main limiting factor for both datacenter and renewable build-outs, hence all the on-site gas generation. Throwing more money at it isn't likely to solve the problem either: the grid expansion is slow for many other reasons.
[3 hidden]5 mins ago
If the gas plant option were banned they’d be forced to build out solar farm capacity.
[3 hidden]5 mins ago
Datacenters run 24/7 so it's pretty hard to make them carbon neutral unless they're somewhere with hydro or nuclear.
[3 hidden]5 mins ago
Building a data center _anywhere_ except right where there is hydro or nuclear seems like a very bad idea
[3 hidden]5 mins ago
How about advanced geothermal?
[3 hidden]5 mins ago
Geothermal is not renewable. Mantle conductivity is terrible.
[3 hidden]5 mins ago
If geothermal isn't renewable, solar and wind is neither. Nor hydro or nuclear
[3 hidden]5 mins ago
No, I mean that once thermal energy is extracted from the crust, it can take quite a long time to be renewed by conduction from the mantle. It's renewable on geological timescales, but not human.
[3 hidden]5 mins ago
That seems like a localized belief.
[3 hidden]5 mins ago
No, it’s not, battery technology exists to solve this problem and is being deployed today at scale.
[3 hidden]5 mins ago
If carbon was expensive enough those data centers would just buy/build tons of batteries (it could even be old lead-acid batteries) and use that for 24/7 running.
[3 hidden]5 mins ago
Batteries don’t replace the need for carbon-free generation if they’re going to run 24/7. Each new power-hungry datacenter would need its own generation infrastructure, even if it’s provided by a local utility company.

So batteries would work if you could pair each datacenter with a large enough area for solar and wind, perhaps, but I don’t know that we have enough unused land area for that. This may be what keeps nuclear around for a little bit longer, high power density is its bread and butter.

[3 hidden]5 mins ago
couldn't you have solar + wind (peaks at day vs peaks at night) plus batteries plus a small peaking gas plant for really bad stretches of weather? On a yearly basis it should be possible to average 80%+ renewable power for it, at the expense of course of using a lot of land and costing way more than coal+gas

I am hoping at some point deep geothermal will help with this as it works anywhere and would provide predictable 24/7/365 base power

[3 hidden]5 mins ago
People already hate data centers going in. Now you're going to expand their footprint 10x or more for the solar and wind farms? Lots of protests for that too I'd imagine.
[3 hidden]5 mins ago
The US uses 12 TWh of electricity per day. Globally, that figure is 60 TWh. Global battery deliveries are about 1TWh per year. Even just building half a day's worth of storage amounts to decades of battery production. Not to mention, it would involve reducing progress towards EV adoption.

The scale of grid storage is truly massive. This is why plans that call for a grid primarily powered by intermittent sources assume some other technology like hydrogen or compressed air delivers storage at a fraction of the cost of existing systems at massive scale.

[3 hidden]5 mins ago
The sun does shine for significantly more than 2% of the day though
[3 hidden]5 mins ago
The capacity factor of solar is 10-25% depending on location. Yes, the sun shines for half the day (on net over the year). But cloud cover diminished that figure significantly.

Worse yet, for many regions solar's worst performing months in the winter is also where energy demand is highest. There's definitely ample room for solar in sunny places like Australia, especially if they gave good hydroelectric potential. But it takes a very specific set of geographic conditions to make it work well with fossil fuel plants to pick up the slack of periods of under production.

[3 hidden]5 mins ago
Only about half, and depends on the weather. You would need a truly massive amount of batteries.
[3 hidden]5 mins ago
Grid scale batteries. Pumped hydro. Molten salt. Etc, etc. It's not easy, but we need to do it.
[3 hidden]5 mins ago
Not we. They. They need to do it. The general public should not be on the hook for anthropic's energy needs. Make them pay for every dollar of power plants and grid upgrades.
[3 hidden]5 mins ago
> The general public should not be on the hook for anthropic's energy needs

Well no, but the general public _are_ on the hook for getting the laws changed.

[3 hidden]5 mins ago
Bullshit. There are plenty of spaces with acres and acres of space in America for solar and batteries. They just don't care.
[3 hidden]5 mins ago
> it's pretty hard to make them carbon neutral

So what? Is getting people to form friendships, even relationships with AI more important than the friggin' environment? Is it so important to have office drones sending emails written by LLMs in order to receive replies from LLMs? Is it so important that everybody can write a TODO list app?

[3 hidden]5 mins ago
I'm not apologizing for them, regulators could make datacenter construction contingent on clean energy sources being available, or at least substantial investment in the grid.
[3 hidden]5 mins ago
There's no moral for some people, is there? You can build data centers as long as you can bribe enough legislators without being found out.
[3 hidden]5 mins ago
If that's the only use you see for LLMs, maybe you're the unable to see what great intelligence can accomplish. Introspect maybe.

About what we just did with the GRH. And what is yet to come!

[3 hidden]5 mins ago
> About what we just did with the GRH

Killing maths studies, and subsequently science and engineering in two or three generations. That's what "we" just did.

[3 hidden]5 mins ago
It's not that they can't, it's that they don't want to.

They go for the cheapest option, even if they've had practically infinite money until now.

[3 hidden]5 mins ago
Did anyone ever think these people cared about anything other then money? Spacex basically operates out of texas so they can power dcs with gas. It's part of why musk dropped half a billion on midterms.

I think all of the major ai labs are running on gas it's ridiculous

Meanwhile Chinas grid is super healthy and has a tremendous amount of renewable energy... This is what happens when you don't pay into infrastructure as a civilization or require corporations to participate or have accountability for the future of the people they serve.

[3 hidden]5 mins ago
China generated 5,500 TWh last year with coal. The largest AI data centers running at full power would use something on the order of 17.5 TWh.

Yes I agree these data centers should be using more renewables, but the China comparison is odd given that they burn enough coal to run 300+ 2 gigawatt data centers.

[3 hidden]5 mins ago
Like half of chinas power generation is coal. And it's a country of 1 billion. Coal is way dirtier than gas.
[3 hidden]5 mins ago
" Spacex basically operates out of texas so they can power dcs with gas."

Texas is also the #1 state in solar and wind production as well as growth rate.

[3 hidden]5 mins ago
Do you know how long it'd take to set up a solar array large enough for even half a typical AI data center?

I doubt Anthropic has that time right now considering the amount of stuff they're actively doing.

Whether or not it's the right thing is a different story, but just imagine the realities and the time sink of actually setting up solar for a data center powerful enough to serve stuff like a frontier model.

[3 hidden]5 mins ago
False dichotomy. The hidden premise is that any of this is necessary.
[3 hidden]5 mins ago
It’s all necessary. It’s a race to profitability before the AI bubble pops and their stock becomes worthless.
[3 hidden]5 mins ago
We could have just stayed in grass huts on the savanna
[3 hidden]5 mins ago
Why would that have been a bad thing?
[3 hidden]5 mins ago
To play devils advocate it’s not a bad thing. It’s actually ideal and 99% of the human experience. People did not choose civilization. Civilization was forced upon people to manage higher density populations. Starvation is worse than civilization.
[3 hidden]5 mins ago
All life expands until it runs out of resources. Fairly sure cyanobacteria killed 99% of life on earth (anaerobic life) at some point. Trees killed 90% of life once. At least we can learn to do better. Hopefully.
[3 hidden]5 mins ago
You can still go back to grass huts if you want. Having the option to not be in one seems preferable to having no option at all though.
[3 hidden]5 mins ago
We could have chosen not to become multi cellular organisms. That was the original mistake.
[3 hidden]5 mins ago
Why does their urgency mean that the rest of us have to suffer?

If I decided to ignore laws/regulations when building/repairing my house, I don't think a court would be interested in "I'm in a rush, so didn't have time to get planning permission etc".

[3 hidden]5 mins ago
> Do you know how long it'd take to set up a solar array large enough for even half a typical AI data center?

Perfect reason to slow down the frontier progress, like Dario, Sam, and Elon all spoke up and said was a good idea...

[3 hidden]5 mins ago
So let’s just burn the planet?
[3 hidden]5 mins ago
The irony is that we've been burning the planet to pursue the advancement of technology and the 'grandness of the human race' since the industrial revolution.

This is, unfortunately, just the next chapter of the same story.

More unfortunately, it feels as if the only way to change the direction of the story is for there to be real, actual, visceral consequences for a majority of the people that are powerful enough to continue pushing in the current direction. Fatally unfortunately, said powerful people will be the last to feel real, actual, visceral consequences. There is a lot of 'worthless human shield' sitting in front of them that has to be ploughed through first.

Doubly ironically, maybe it's best if AI does destroy humanity and let the Earth find a more sustainable balance of life.

It will be interesting to see, if AI does destroy humanity, will it let the Earth find a more sustainable balance of life, or will it accelerate the domination of the natural world in pursuit of its inscrutable goals? (likely to its own suicidal detriment, given it is based on learning from humans)

[3 hidden]5 mins ago
of course; the realities are that gas and coal are heavily subsidized, so theres never been a chance of making a fair comparison among real dollar per megawatt

the real bill comes due in about 4-5 years :)

[3 hidden]5 mins ago
>Anthropic can also do at least something

No, because solar panels are notoriously hard to scale, especially in high density applications like datacenters. Otherwise it's like saying "my calculator runs off solar power, so why can't they make a car run on solar power? Clearly it's because Ford/BYD not giving a f"

[3 hidden]5 mins ago
The investments in AI is in the trillion of dollars, surely that's enough capital to figure this out. It might require inventing or improving existing technology.

Before you respond "it will be too slow", the urgency here is entirely artificial – there's no reason we need to be going this fast.

[3 hidden]5 mins ago
> It might require inventing or improving existing technology.

Or it might result in investing trillions of dollars and decades only to discover that there are only marginal gains to be had. There is no guarantee that throwing money at Problem X results in any kind of solution or that solutions to specific problems are even possible.

This is aside from the sibling commenters' very good point that this view assumes there are no concrete benefits or advancements to be made that depend on AI, which seems very unlikely at this point. Unfortunately, with or without AI, the problems we're facing socially, ecologically, medically, and scientifically are increasing. We can either use AI as a tool in the hopes we can at least delay the most serious consequences or just let them overtake us. Fossil fuels and climate change are just one of a million very pressing issues, despite the over representation it gets, and it seems highly likely that the benefits here exceed the costs. Addressing a downthread comment about labs saying we need to "slow down" - they are saying this for self-serving reasons and in their quest to ensure AI stays gatekept, not because they are worried about environmental impacts.

[3 hidden]5 mins ago
I don't see how we could invest trillions of dollars and get nothing to show for it. Consider the Apollo and Manhattan projects, both were large investments that resulted in significant scientific and engineering progress, why would this be different?
[3 hidden]5 mins ago
Because physics and materials science are ultimately subject to reality, they are not a slot machine you put money into and eventually get a payout.

Even your own examples tend to show that. There were in both cases rapid advances followed by diminishing returns as all low-hanging fruit was worked out. 50 years later space travel is still very expensive and difficult. Eventually we may attain reasonably economical orbital travel with sufficient advances in rocketry and materials science but there is no reason to believe any amount of money and time will, for example, give us FTL spaceships.

[3 hidden]5 mins ago
Contrary to popular beliefs, spending money doesn’t actually grant you godlike abilities such as defying physics and chemistry.
[3 hidden]5 mins ago
I'm no physicists, but we sure seem far from the limits, we haven't even figured out fusion yet and there are physicists, who presumably know what they are doing, working on that.

It's not like we have exhausted all avenues and are stuck. Solar panels are getting cheaper and more efficient, storage is a problem, but it has not seen that much investment. More nimble forms of nuclear fission, like SMRs, is an active area of research. As mentioned above, we are making progress on fusion.

The idea that we are anywhere close to the diminishing returns on investment in these areas seems completely wrong to me.

EDIT: The dichotomy in this discussion is striking: we need to disregard almost all negative externalities on speculative investment in AI, but speculative investment in green energy is not worth it because it might not yield sufficient results.

[3 hidden]5 mins ago
Then why does the AI industry exist? The entire premise of this horse shit is that if we spend enough money, we can summon a demon to destroy humans
[3 hidden]5 mins ago
> There is no guarantee that throwing money at Problem X results in any kind of solution or that solutions to specific problems are even possible.

And yet, we are currently sacrificing our entire economy, our social cohesion, most of the world’s productive resources, and the careers and futures of thousands to millions being displaced, to chase the possibility that if we computer hard enough, computer will be alive

[3 hidden]5 mins ago
throwing large amounts of money on llms is proof that throwing large amounts of money on any problem will lead to some form of acceptable solution
[3 hidden]5 mins ago
> urgency here is entirely artificial – there's no reason we need to be going this fast.

What makes you believe that time is irrelevant? If AI allows for breakthroughs in mathematics, science, engineering, computing, and health care—and it does seem to be delivering on this promise so far—what in your estimation is the net cost to humankind of delaying those breakthroughs by even a year? It's not zero, is it?

[3 hidden]5 mins ago
I mean sure, but the labs themselves are saying we need to slow down. Your position is only reasonable if the negative externalities are outweighed by the positives. Not to mention p(doom), which – unless yours is 0% – should mean we must slow down or stop immediately. We're justifying definitive downsides with speculative upsides.

More than anything though, it doesn't matter what we in tech think because we have failed to convince the general public that any of this is a good idea.

[3 hidden]5 mins ago
Let's hope AI can solve climate change before 1.5 degrees is inevitable...

Well done humans, well done. Slow clap.

[3 hidden]5 mins ago
> before 1.5 degrees is inevitable

You're expecting people to get working time travel from LLMs ?

[3 hidden]5 mins ago
(that was my point)
[3 hidden]5 mins ago
> If AI allows for breakthroughs in mathematics, science, engineering, computing, and health care.

Math, maybe. Science, where? Computing, anything measurable? Health care, where?

FYI, pre LLM techs don't count. Ain't nobody building $3tn worth of DCs for AlphaFold.

[3 hidden]5 mins ago
Solar power doesn’t work at night, and AI data centres are a 24/7 load.
[3 hidden]5 mins ago
I'm not sure if the car analogy applies. Data center solar power can scale without the requirement of being portable.

The problem seems to be that they want the power online quickly and they can't get solar quickly. The article mentions the inefficiency of their generators. It is costing them quite a bit extra to generate power like that, but that's their only option of they want the power immediately.

It would be nuts to do this long term. You might argue that they don't give a f about the environment , but most people will agree that they give a f about cutting costs.

[3 hidden]5 mins ago
> Data center solar power can scale without the requirement of being portable.

You just have other issues like needing to get transmission and storage capacity built, which takes forever because of planning review, environmental assessments and NIMBYs.

[3 hidden]5 mins ago
Not the best example, surely? My car runs on solar power (gathered from my roof)
[3 hidden]5 mins ago
You can, but how many cars are actually powered this way? Same goes for datacenters. It's not like electrons from solar panels physically don't work with GPUs, it's just that it's hard to make it work due to other constraints.
[3 hidden]5 mins ago
For electric cars it isn’t uncommon for people to put solar panels in their homes or their garages to charge them. I think the car themselves with solar panels won’t work because most people store the cars in garages usually, away from the sun - it is usually very harsh on everything that isn’t solar panels.
[3 hidden]5 mins ago
> how many cars are actually powered this way?

Some. Hence "why can't they make a car run on solar power" isn't the obviously dumb question you make it out to be.

[3 hidden]5 mins ago
1. From a pure argument lawyering point of view, my original question was "why can't [Ford/BYD] make a car run on solar power?" with comparison to a calculator, which implies a car with built-in solar panel. Otherwise the statement of "they make [thing] run on solar power" becomes meaningless. eg. "Apple made a macbook run on solar power" (because you can take a regular macbook and plug it into any solar panel.

2. There probably are non-zero amount of datacenters out there that runs off solar power. It's just not commonly practical, just like with electric cars.

3. The main point of the comment is that scaling challenges exist, and you can't take a residential use case and extend it to a datacenter. The same applies for a car, even if you could do it. A solar powered calculator is easy, just slap a tiny solar strip and it just works. A solar powered car is far harder to pull off. If you don't work from home for instance, you'll need a battery system to store the energy when you're not home.

[3 hidden]5 mins ago
Are you intentionally trying hard not to understand?
[3 hidden]5 mins ago
This is simply an argument against datacenters. You're pointing out a negative efficiency, and the ostensible point of concentrating data processing is to produce savings.

If you can't figure out how to get renewable power to a datacenter, but you can easily figure out how to get it to the consumers of a datacenter, and you're talking about a technology that scales down to small groups well and individuals only a bit less well (but with constantly working agents utilizing compute, better than we once imagined), what's the reasoning behind datacenters at all?

If datacenters can make a case for massive increase of demand in the future, they're simultaneously making a case that individuals will have enough demand to saturate AI located closer to them or on premises. If anything, that should be a call for p2p software to help people share or borrow excess compute with and from their neighbors in exactly the same way that home solar panels use the grid.

Datacenters are also central points of failure. Distribution is resilient.

[3 hidden]5 mins ago
Sounds like we shouldn't be building big data centers if there is only nonviable means to power them then?

I know meta and their ilk are trying to cut deal for nuclear power. I'm sure we can imagine they will take safety, the environment, etc very seriously too.

[3 hidden]5 mins ago
>because solar panels are notoriously hard to scale

Are you high?

[3 hidden]5 mins ago
Yeah also they are loss making and will basically be fighting for survival now that they're public and visibly not profitable. Private investors have offloaded to the hyped public bag holders and now we're seeing how the sausage is made.
[3 hidden]5 mins ago
Rules are for peasants.
[3 hidden]5 mins ago
You thought they gave a fuck after they stole the collective works of humanity then sold it back to us? These companies have always been dens for the devil to do their bidding.
[3 hidden]5 mins ago
YOU make a poor argument. Most sysadmins read Slashdot all day, and complain about bloatware or Windows 27 Server security profiles. By default, is your AI research into latent atomic arguments... or monotonic parabolas? My policy lawyer sniffed glue as a child so you can't convince him. /gosh wow
[3 hidden]5 mins ago
I will be a bit provocative here, but if you do want Green (or at least Greener) A.I....Just use Mistral and the Chonk.

- French Nuclear (where Mistral is) is around 5 gCO2/kwh

- Gas plants (where all US models are) are around 300-350 gCO2/kwh

- Coal plants (where most Chinese models are) are around 800-900 gCO2/kwh

So in term of consumption, your queries are around 65x greener in term of energy consumption hitting a French datacenter than a US one.

Why US data providers are not building more Nuclear power ?

Answer: Time to market. It takes less than a year to setup a 1GW gas turbine. And close to 10y to setup a similar capacity in Nuclear.

Why France does not have the problem ?

Answer: Because they do have a ridiculous oversized Nuclear park with 150 TWH of overcapacity per year.

The only other areas of the world that can afford to have both the grid power necessary for A.I datacenters and the low Carbon intensity are: Sweden, Norway and Canada hydro (in some areas).

[3 hidden]5 mins ago
And Melenchon wants to phase out nuclear. I don't envy French the choice they'll have in next elections.
[3 hidden]5 mins ago
And I think you describe here why American A.I actors did not rush to build Datacenters in France even with such a low Carbon electricity (with the exception of Google, and I believe, Meta).

The political landscape right now is such a shit show that it is a huge risk on investment that they are probably not willing to take.

There is a slight chance that this get significantly better if Lisnard get elected.. But that is a pretty small hope for now: Polymarket evaluate the probability only to 17-18%

[3 hidden]5 mins ago
It is much worse than I thought. I’m a power user. I have some reflections to make.

Hope for solar datacenters in the Middle East?

[3 hidden]5 mins ago
The Australia–Asia Power Link (AAPowerLink), aka SunCable, is a proposed undersea HVDC cable to export solar power from northern Australia to Singapore via Indonesia. Due to doubts about the economic feasibility of constructing the cable, and question marks over the willingness of Indonesian and Singaporean governments to approve it, they've been exploring pivoting to building solar-powered AI data centres in northern Australia instead.
[3 hidden]5 mins ago
I somehow suspect that datacenters are only economically viable if you run them round the clock, right? So at night they'd still be using gas. I guess the cost of compute would have to come down a lot, in order to power down a DC and spin up one where sun is shining instead.

Although I guess for current AI workloads, it could be viable, since the additional latency of sending your (small, text only) prompt around the world for processing is little compared to the time taken to process (whereas for other stuff being close to users matters more)

[3 hidden]5 mins ago
> I somehow suspect that datacenters are only economically viable if you run them round the clock, right? So at night they'd still be using gas.

You could build batteries, charge the batteries during the day, run the DCs off it at night.

SunCable / AAPowerLink is likely going to be connected to the Darwin grid, which is currently not connected to the main East Australian grid; but building a connection to the main Australian grid would be a lot cheaper than undersea HVDC to Singapore. Once that happens, there would also be the potential to use pumped hydro (including Snowy 2.0) for storage.

[3 hidden]5 mins ago
You could use batteries, but batteries are very expensive per amortized Mwh. It'd be more expensive than just running natural gas turbines 24/7.
[3 hidden]5 mins ago
solar during day plus battery park, just like factorio
[3 hidden]5 mins ago
> datacenters in the Middle East

Given this year’s events [0], even Sahara seems more likely.

But you’d still have to build it out from ≈scratch, including cooling, internet connection infrastructure, and maintenance, so I’m not sure what the time horizon on amortization would look like, even with the cheap solar power.

[0] https://www.reuters.com/world/middle-east/amazon-cloud-unit-...

[3 hidden]5 mins ago
Aside from a dirt road from the nearest port and a few fibres to various IXPd what do you need to build in Morocco or Egypt that you don’t need elsewhere?

Obviously security is a concern which limits the number of locations, but there’s still plenty of places with stability, security, land, and the lack of taxation and tarrifs that the US puts on energy.

[3 hidden]5 mins ago
There have been many proposals to build solar in Africa for exports to Europe, but none of them has gone anywhere. That's because the cost of capital is too high. It's cheaper to just build in Spain.

Globally, places like Australia and India should be more competitive for solar than the Sahara, although both Australia and India have a lot of dirty power generation to replace as well.

[3 hidden]5 mins ago
> It's cheaper to just build in Spain.

I find the concept of agrivoltaics really exciting for climates such as found in Spain.

https://www.google.com/search?q=agrovoltaics+in+spain+resear...

[3 hidden]5 mins ago
I recommend anywhere else. Datacenters in the Middle East will use fossils no matter what promises are made for very obvious reasons (economic and otherwise).
[3 hidden]5 mins ago
They could run on solar in US throughout the daylight hours. Just make the tokens cost 5x after the sun sets That would reduce the overnight demand enough the data centers could run through night on batteries.
[3 hidden]5 mins ago
You can find inference providers that are renewable powered
[3 hidden]5 mins ago
Do our expert controls allow top of the line chips in those countries?
[3 hidden]5 mins ago
There's always an option of... wait for it... datacenters in space.

*runs*

But then of course that's another front-page submission with its own long thread of commenters feeling smart for calling it stupid, despite recently becoming a worked out plan and addressing exactly the (subset of reasonable) things people are complaining about here. In fact, we had those a week or two ago.

I swear the tech space turned into thought-space NIMBYs. "I don't understand any of it, but something about it pattern-matches to something I don't like, therefore it's stupid and people behind it are evil". In this thread, and in every other thread recently, too.

[3 hidden]5 mins ago
I looked into it just now, and it appears SpaceX is targeting deployment of the first satellites in 2027.

To deploy 1 GW of capacity, apparently 100 Starship launches would be needed.

On earth, 23 GW are currently under construction, 100-132 GW is already active, and about 100 GW should be added by 2030.

SpaceX's CFO supposedly cited 2029 as the timeline for multiple GW of compute in operation in space. Musk's vision seems to be launching 100GW/year or 1 million ton of satellites.

This would require 10k launches of Starship per year under ideal conditions, or 5k if Starship Block 4 can double the payload capacity to 200t as planned. This would mean 14 launches per day.

It does sound like a lot. Starbase Louisiana, at a $100B buildout, is planned to have 12 launch pads that could handle 30 launches per day.

Obviously there's much more needed than some launch pads. Anyway, that's the numbers and projections I found.

[3 hidden]5 mins ago
Data centers in space won’t help with CO2 impact… at ~3,000 tons of CO2 per starship flight.
[3 hidden]5 mins ago
Apparently that's still 30x-150x less than gas or coal over 5 years.

It might shrink to 3x-10x when considering secondary atmospheric effects, such as on reentry, if they don't use 'graveyard orbits' instead.

[3 hidden]5 mins ago
And that’s the optimistic take, assuming Musk’s projections are correct. Yet they are so often wrong there’s a full Wikipedia page just for things he said would be coming (often “next year”) and didn’t.
[3 hidden]5 mins ago
Yes, and I am curious about the financial viability of that compute.

From a capacity perspective, if SpaceX remains successful in scaling up their Starship rocket, it seems putting AI compute into space could at least realistically happen at significant scale.

[3 hidden]5 mins ago
The amount of CO2 spent launching the data center to space probably offsets the lifetime saving of the solarbased energy used by the data center. That is ignoring filling atmosphere with toxing fumes on re-entry, and that after the short lifetime in space the servers, GPUs and solar panels don't find a second life.

Should SpaceX solve the problem of requiring fossil fuels to launch (doable) and the burning things on re-entry (harder) I don't oppose solar data centers. But without those, it's just externalizing the environmental cost to future generations.

[3 hidden]5 mins ago
Still seems insane attempt to leap over a lot of development that space launch improvements finally enable.

Instead we should be setting up in space infrastructure to enable manufacturing and most importantly space resource use.

Then we can build anything more cheaply at a much bigger scale.

Doing space compute clusters right now is like trying to do regular trans-atlantic airline service with biplanes with floats. While that was possible and done to a degree, it all only took off with more advanced airplanes and supporting infrastructure.

[3 hidden]5 mins ago
> Doing space compute clusters right now is like trying to do regular trans-atlantic airline service with biplanes with floats. While that was possible and done to a degree, it all only took off with more advanced airplanes and supporting infrastructure.

Same might be the case with space. Building infrastructure to enable cheaper infrastructure to enable massively cheaper operations is usually the right approach on paper, but also usually near impossible to fund directly. What happens instead, is attempts at doing something directly, which predictably fail but prove the viability of the idea and justify marginal investment in the infrastructure; couple rounds of that, and in a roundabout way, we eventually arrive where we could've arrived 10x cheaper if the economy could fund long-term plans. Alas.

[3 hidden]5 mins ago
> "I don't understand any of it, but it pattern-matches to something I don't like, therefore it's stupid and people behind it are evil".

This is pretty bad-faith. What is the point of your comment? You’re sowing hate and attacking a nondescript group, for what? What does that achieve apart from being hateful?

Yours is one of the few usernames I recognise on HN. I remember seeing some comment from yours (I think it involved Star Trek) which I found interesting at the time and somehow the username stuck with me. Since then, every time I bump into your comments they have gotten more hateful and generally worse and more tribalistic takes.

It genuinely saddens me. Especially as you have in your bio “I'm constantly looking for opportunities to do something actually useful for humanity”, but your words reveal the opposite.

I’m not going to be specific about past posts because frankly I no longer remember their content, only the disappointment I felt. But I can take this one as an example. In addition to every thing I mentioned, I find it a pretty bad take to claim that adding more shit to orbit improves humanity. We need less crap up there, not more. I also find it nonsensical to defend (if we can call it that; you just attacked detractors with no explanation) something that is purely theoretical (which it objectively is; it doesn’t exist yet).

Anyone who has paid attention in the last decades will see all the unkept promises. You can see them in this very submission. Blockchain and AI were full of proponents making the argument that “all this energy consumption is good, actually, because it’ll increase investment in renewables”, but obviously what we’ve seen is an even bigger spend on fossil fuels which are exacerbating the problem. Believing the next promise from proven liars who make everything worse every time is not a smart strategy.

Anyway, I’m not entirely sure why I felt I should write to you specifically. I guess I once saw the possibility of something better and am hopeful that being respectfully called out might help you to reflect. Those of us capable of introspection (which I believe includes you) sometimes need a little nudge to get back on track to the person we want to be. I know that’s true for me. Sometimes I help others, other times they help me. I hope this helps you.

[3 hidden]5 mins ago
You win ironic post of 2026.

I know there are smart people on HN, and I don't believe most of what (person that shouldn't be named) says. But I agree with OP, the amount of hate anytime the topic comes up is frustrating. And the arm-chair aerospace engineers are are the most annoying group.

[3 hidden]5 mins ago
I’m not sure you understand the meaning of irony. But I can say for sure you didn’t understand my post, because it’s exceedingly clear you missed the point. Data centres (whether in space or on Earth) and the people who oppose or defend them are very much not the focus. Fortunately, I believe the person it was meant to might understand. That comment had an audience of one.

Of course, comments and downvotes like yours were very much expected. On HN I’ve come to anticipate lack of nuance and the belief any disagreement or criticism is hateful and looking for a fight.

I can only try to counteract that one post at a time, though I can’t say I’m feeling hopeful that the bother is worth it.

[3 hidden]5 mins ago
Keep going.
[3 hidden]5 mins ago
Actual order of events:

1) Musk retches up another one of his Hyperloop-style dumb-ass ideas whilst patting himself vigorously on the back

2) His companies grovel hilariously to backfill the reasoning for why this is a Good and Smart idea, actually, while actual experts put on a wtf face (“worked out plan” goes here)

<this is where billions come in handy, by the way, since you can just make people work on whatever stupid shit you want without getting any buy-in from the sane>

3) The technorati follow the their glorious thought leader off his cliff

[3 hidden]5 mins ago
Hope for better regulations to set incentives?
[3 hidden]5 mins ago
I do not think that US companies will deem the Middle East as a safe place for building DCs
[3 hidden]5 mins ago
Nuclear power is the future.
[3 hidden]5 mins ago
In some alternate universe where cost doesn’t matter, sure.
[3 hidden]5 mins ago
Sweet! Power in 2046 at the earliest of we start today

That is simply not workable, and the suggestion needs to die because it sucks motivation from solutions that can work in a year or two.

[3 hidden]5 mins ago
It taking this long is more of a US problem, you can look at China, Japan, and South Korea as counterexamples of when the bureaucracy isn't broken.
[3 hidden]5 mins ago
No Western country has ever built a nuke in less than 20 years.
[3 hidden]5 mins ago
You mean recently? Western countries have not had a nuclear energy plan over the previous years, many of them wanted to phase out nuclear reactors, some countries even started to do it, and some entirely did it. It's not a very encouraging situation to build nuclear reactors quickly. Countries which made ambitious plans with nuclear reactors in the past managed to build dozens in less than 20-30 years. But that does require massive investments. It's much easier to invest less money in solar panels and wind turbines, whether that second choice allows to remove fossil fuels as fast is something we can already observe.
[3 hidden]5 mins ago
Please provide a source for a western country that built a nuke in less than 20 years from project start to power into the grid.
[3 hidden]5 mins ago
https://en.wikipedia.org/wiki/List_of_commercial_nuclear_rea...

You can look France, Canada, USA. For example Civaux 1: they started to build it in 1988 and in 1999 it was producing electricity: https://world-nuclear.org/nuclear-reactor-database/details/C...

I plotted how fast nuclear reactors have been built in the world based on this page. It starts at 5 years, peaks at 7 years, and ~90% of reactors are built in <15 years. The last reactors built in western countries (which took a long time to be built) don't represent the average situation. It's taking the worst scenario, which is very pessimistic (why?).

Of course you can include what you want in "from project start", maybe you're reffering to when a politician says "we should build it" rather than when the construction began. In many western countries projects for future reactors have already started (UK, France etc.). A large country won't build 20 reactors in 7 years starting tomorrow, but if it starts building all reactors simultaneously it can build 50 reactors in <30 years.

Maybe what you're also saying is western countries take too long between the decision to build a reactor, and actually building it. I agree with you that we should remove a lot of steps and discussions to accelerate these projects.

[3 hidden]5 mins ago
France's historical fleet of plants took 5 years on average to build. They are all totally safe. The inefficiency in the modern day is completely self inflicted and voluntary.
[3 hidden]5 mins ago
You're not wrong but that's not exactly fair either.

France did build out a lot of them over a short period of time. This worked because they standardized the entire thing. They then steamrolled all local objection in order to build them quickly.

So at that point, the design was done, they had factories set up to build the parts and full backing of the national government allowing them to push this through.

So yes, the Nth reactor like that might be built in under 5 years because the design was already done and they built many. This is just not where we are today

[3 hidden]5 mins ago
I asked ChatGPT, and it referenced Plant Vogtle Unit 3 in Georgia, which took 17 years.
[3 hidden]5 mins ago
France's latest was
[3 hidden]5 mins ago
Long term plan is bad because it will suck motivation from short term plan? You sound like someone who cares a lot about quarterly returns
[3 hidden]5 mins ago
Long term plan is so long and so crushingly expensive it’s not a workable solution to the needs of today.

Short term plan is much faster and cheaper. Therefore it’s a much better plan.

China add an entire US grid worth (1,281GW) of renewables every 3 to 5.3 years. Why wouldn’t the US do the same and have the problem well in hand in 5 - 10 years rather than waiting 20+ for a single nuke?

[3 hidden]5 mins ago
Don't ban datacenters, require new 110% renewable energy capacity for new datacenters
[3 hidden]5 mins ago
Just price greenhouse gas emissions (or fossil fuel extraction). These sort of weird hoops for new users to jump through fail to consider the possibility that some things are worth doing even if they increase net emissions. A price on emissions would move savings to the least valuable emitters, rather than the newest emitters.
[3 hidden]5 mins ago
Republicans just took an all time shellacking because of energy prices. Carbon pricing is so far out of the realm of political possibility right now it barely makes sense to acknowledge
[3 hidden]5 mins ago
Carbon tax proposals are usually designed to be revenue neutral.
[3 hidden]5 mins ago
Try explaining that to voters
[3 hidden]5 mins ago
The problem is then voters go turbo Nazi over gas prices
[3 hidden]5 mins ago
All that means is the price of generation is bid up, making it more costly for everybody else. For example, PSE in Washington State has a mandate to be 100% renewable in ~fifteen years. They recently got outbid for a solar plant in Oregon by a compute company (Amazon? can't remember off the top of my head).

So PSE's consumer rates are scheduled to go up 30% just in the next three years. It's not going to slow down if AI is bidding up every power plant. And in states without the mandate, it just makes keeping gas plants running more cost-effective, because power is fungible, so there's still a net increase in fossil fuel generation despite said mandate.

[3 hidden]5 mins ago
Great idea, when will you run for office to get this bill passed?
[3 hidden]5 mins ago
The builders of these datacenters will tell you that those two things are the same thing. If you don't let them build datacenters without any regulation, in the cheapest possible spot, with the dirtiest possible power, then you are in fact banning them.

This is pretty standard capitalist propaganda.

[3 hidden]5 mins ago
Yes but anti-caps focus way too much on armchair analysis, the dont's and bans and forecasts of failure. They're not completely wrong, but to move the needle they need to propose actual solutions and get them to stick - even if incremental.
[3 hidden]5 mins ago
You'd think people in the position to call the shots aren't all 3 year old kids.

- "But I don't want this toy here!"

- "It must go here, because over there is a big crocodile that will eat my toy."

- "Oh okay then..."

Seriously, your "standard capitalist propaganda" as stated is a 3-year-old argument. Any reasonable adult will just say "then don't build it".

[3 hidden]5 mins ago
For which there would be all the money you could want, but why let morality come in the way of more profits
[3 hidden]5 mins ago
Serious question: can the US people afford that?

At least from the news I've seen across the ocean, I was under the impression that the USA's grid was already stretched way too thin because of their lack of investment in renewables, and many people are against data centers because electricity is already too expensive for normal people. Or was that just news cycle hyperbole?

[3 hidden]5 mins ago
US power prices are still much, much lower than EU.

The EU averages 0.28 eur/kwhr The US pays 0.17 eur/kwhr

The problems are generally exacerbated by renewables. e.g. Transmission issues to do variability, lack of storage, grid sync.

The US's problem is that technological advancements generally reduced usage. So the total change in demand was flat from ~2005 onwards. The utility companies haven't build new capacity for about 20 years.

https://ec.europa.eu/eurostat/statistics-explained/index.php...

https://fred.stlouisfed.org/series/APU000072610

https://www.eia.gov/todayinenergy/detail.php?id=67284

[3 hidden]5 mins ago
The US people have nothing to do with that. The public US grid is indeed not sufficient, but if you read the article, you'll see that these companies now bring their own power. And since fossils (primarily gas/methane) is the only power source that can deliver huge amounts of electric power in a foreseeable time frame, companies like Anthropic are primarily interested in that. You can buy a couple hundred gas generators, stick them in containers next to an oil field in Texas (which would otherwise just burn the gas) and have a GW for your datacenter before you'll even have the permits for a conventional fossil or nuclear power plant or an equivalent solar/wind farm.
[3 hidden]5 mins ago
If the gas is getting burned either way, why not do some useful work with it?
[3 hidden]5 mins ago
That's exactly what the data center developers discovered. If the oil producers hadn't been mindlessly blowing that shit in the air or burning it, there wouldn't be a data center boom, because there wouldn't be an immense untapped local power source across many desolate places in the US. You can't easily move that power to population centers, but you can simply move your data centers there.
[3 hidden]5 mins ago
Why would they not bottle up this gas and sell it themselves?
[3 hidden]5 mins ago
Because transportation costs money. It's the same reason why farmers occasionally dump milk or let crops rot on the field because it's not worth the money to harvest/distribute.
[3 hidden]5 mins ago
Solar + battery is perfectly capable of delivering the power required; it would just cost more up front, so they don't want to do it.
[3 hidden]5 mins ago
No it isn't. Not on these scales. A solar farm capable of producing 1 GW average power over a 24h period with battery storage would need 100-200 km^2. This kind of power production simply can't be delivered in remotely comparable time with solar.
[3 hidden]5 mins ago
You’re getting downvoted, but you’re right
[3 hidden]5 mins ago
> USA's grid was already stretched way too thin because of their lack of investment in renewables

That doesn't follow though. If USA's grid is stretched too thin, it's because of lack of investment in power generation. Not renewables specifically. In fact, renewables are still the most roundabout/difficult way to go about addressing that. And as unfortunate as it is that gas is the economically-best option (I'd vastly prefer small nuclear), if data centers are forcing investment in new power generation, that benefits the grid.

[3 hidden]5 mins ago
> And as unfortunate as it is that gas is the economically-best option (I'd vastly prefer small nuclear), if data centers are forcing investment in new power generation, that benefits the grid.

Even if so, kragen refutes it in a sibling comment, it's only because the negative externalities aren't priced in in the US. Slap a $100/ton carbon tax on it and that problem is solved.

[3 hidden]5 mins ago
No, it’s a lack of investment in distribution, not generation. There’s a huge amount of stranded renewable generation that currently can’t connect to the grid.
[3 hidden]5 mins ago
Renewables generally have highly variable output, which necessitates overbuilding generation capacity across geographically distributed sites, as well as a much larger transmission capacity buildout to balance regional variations in generation while maintaining acceptable supply stability across the wider grid.
[3 hidden]5 mins ago
Is gas really the economically-best option?

CME's Henry Hub natural gas futures NGF27 https://www.cmegroup.com/markets/energy/natural-gas/natural-... are currently at 3.791 in "U.S. dollars and cents per MMBtu", which I take to mean million BTUs. Minimum price fluctuation is given as "0.001 per MMBtu = $10.00", which I guess is because the contract unit is 10,000 MMBtu.

I think the containerized gas turbines for the data centers we're talking about are simple-cycle gas turbines ("SCGT") which are typically 35–40% efficient (https://en.wikipedia.org/wiki/Simple_cycle_combustion_turbin...) using the lower heating value ("LHV") of the fuel, which is about 90% of the higher heating value ("HHV") by which it's priced on the CME. So figure that you get about 33% of those million BTUs in electricity out of the gas generator.

A million BTU is, in modern units, about 1.06 gigajoules. So, just for the fuel, you're paying about US$10.90 per gigajoule. In the folk units used for selling electrical power, that's about US$39 per megawatt hour. That doesn't count the interest and depreciation on the generators or balance-of-plant equipment. (But those costs are relatively small, which is the main reason they're using gas turbines in the first place. https://www.eia.gov/analysis/studies/powerplants/capitalcost... has a capital-cost study of such a simple-cycle gas-turbine plant, with the capital cost coming out to US$713 per peak kilowatt, of which about a third was "Mechanical – Major Equipment". If we assume a capacity factor of 80%, that's US$891/kW delivered. At an interest rate of 6% per year, assuming no depreciation, that's US$53/year/kW. At 8.77 MWh/year/kW, that's another US$6.05/MWh. Depreciation might double or triple that.)

If we compare that to USA photovoltaic PPA pricing in recent years in the https://emp.lbl.gov/pv-ppa-prices we can see that nearly all PV PPAs are cheaper than just the gas you need to fuel your gas power plant. A few are more expensive. The cheapest one this year was US$22/MWh. This includes not just the PV modules and balance of plant but the generator operator's profit as well.

So, I don't think it's correct to say that "gas is the economically-best option" in general. Generally speaking, electrical energy from gas is more expensive than solar energy, though there are exceptions. The great benefit of gas is that the upfront investment is very low (much lower than other fossil fuels, and sometimes even slightly lower than utility-scale solar), and it's portable. There's room in an urban core next to a datacenter for a container full of gas turbines. There isn't room for a solar farm there. You have to site the solar farm somewhere relatively far away and run transmission lines, which involves politics.

It doesn't have to be that way. As Rui Ma explained last year in https://xxcancel.com/ruima/status/1955372325970514161, in China:

> Energy is considered a solved problem. The Chinese government’s investment in sustainable energy — from advanced hydropower to next-generation nuclear — means that, relative to many other markets, electricity supply is secure and inexpensive. Everywhere we went, people treated energy availability as a given. This is a stark contrast to the U.S., where AI growth is increasingly tied to debates over data center power consumption and grid limitations.

[3 hidden]5 mins ago
> I'd vastly prefer small nuclear

Yeah I’d prefer various fantasies that don’t exist too.

[3 hidden]5 mins ago
China has a small nuclear plant going online in January. Others are 4-6 years away.

Not fantasy.

[3 hidden]5 mins ago
It's a cool idea, but even if they appear in 4-6 years, there will be many companies wanting the extra energy during those 4-6 years. And it will be a massive amount.
[3 hidden]5 mins ago
I’m all for emissions regulations. Slow them down until they either implement carbon capture or hook into nuclear.
[3 hidden]5 mins ago
There are five grids in the US, and they're almost compatible in the sense that they all run at the same frequency but they're not in sync. It's not really true that 'the grid' is spread too thin, but it is true that a spike in demand or a drop in supply isn't easily fixed like it is in other countries.

https://www.energy.gov/oe/learn-more-about-interconnections

[3 hidden]5 mins ago
The article is about Anthropic investing in generating more power for itself.

> and many people are against data centers because electricity is already too expensive for normal people

Yeah, exactly. That's why Anthropic is generating more power for itself...

[3 hidden]5 mins ago
That’s all true, the grid has a huge amount of deferred maintenance, and that’s why a lot of data centers are powered by natural gas — because in some cases, you can buy a natural gas turbine to generate your own electricity faster than you can get a grid connection.

If we had rational, sober governance, we would just make the people who want to build data centers pay for the grid upgrades needed to connect both the DCs and the large renewables projects waiting in the interconnection queue.

The problem is that the US is ideologically allergic to central planning, and instead we’re just allowing the money to find the shortest path to power, like a slime mold.

[3 hidden]5 mins ago
> lack of investment in renewables

That's indeed a way to put the active dismantlement that happened.

[3 hidden]5 mins ago
Investing in renewables won't help the grid, we need large scale storage first.
[3 hidden]5 mins ago
news cycle hyperbole...

More than 1/2 of my energy generation at my home is grid sourced as renewable. I have a HA integration that tracks this semi-realtime. The cost is ~14c/kw ... about half of what the EU pays (on avg)

[3 hidden]5 mins ago
What you've described matches some places in the US. It also matches some places in Europe.

But, for counterexample, my power bill in the US is included with my water, sewer, irrigation, and trash collection. Monthly total is between less than 1% ($100) in the summer to around 3% ($300) of a median monthly household income ($10,000) in a metropolitan of about 1.5 million people. Here, the electric grid is owned by the city, not privately operated.

Generally speaking, bigger cities have a high cost of living. And the same can be said about bigger cities in Europe.

But having lived in both places, I find that life is so much harder in Europe, in terms of cost vs income, and in what your money can buy.

Is the media lying to you? It's stretching to focus on specific examples. Sensationalism is profitable. And often times media is colored by corporate or government interests.

[3 hidden]5 mins ago
> We follow the air permitting process everywhere we operate

That doesn't mean much if they build in places with loose regulations.

[3 hidden]5 mins ago
The EPA needs to be restored to enforcing capacity and federal regulations in place. Having per state air pollution laws doesn't make any sense.
[3 hidden]5 mins ago
Isn’t that the whole country with trump 2?
[3 hidden]5 mins ago
States, counties, and cities can have their own laws and regulations.
[3 hidden]5 mins ago
I'm not opposed to data centres in general, but it's insane that there isn't a renewable energy requirement here. Mind you, a lot of this is downstream of America's refusal to transition away from fossil fuels.
[3 hidden]5 mins ago
Given the very limited impact a small amount of latency would have, is there a reason not to build data centers in the Sahara, surrounded by solar panels with energy storage for night use and using a closed-loop water system to sink waste heat deep into the ground?
[3 hidden]5 mins ago
Horrible idea. The heat cooling itself would require more energy. I really don't know why there seems to be an obsession with solar power on HN, if they were truly the panacea for all of our energy problems they would be everywhere but they are not for obvious reasons.

Energy density is completely ignored when talking about why fossil fuel is so popular. You can transport it anywhere cheaply, store it, and it can output when you need it in any condition.

Solar panels themselves lead to ton of pollution when they are scrapped and often abandoned once they are installed leading ton of ecological problems for wilderness.

[3 hidden]5 mins ago
[3 hidden]5 mins ago
I live in Flagstaff Arizona and we had a community meeting on data centers that was really interesting last Wednesday; the organizers have a web site https://noazai.org/

I am slightly negative on data centers, mostly because 90% of the power and environmental problems stem from power generation, not the data centers themselves (according to a speaker at the meeting).

If our country got its act together and fixed infrastructure problems, then I would be more for accelerated data centers. My preference would be slowly developing data centers for 5 to 10 years, get all of our infrastructure in good shape, then do a more robust data center buildout in the future.

The problem, really, is that the elite economic class makes too much profit on endless wars, 750 foreign military bases, etc.

We need to somehow make doing the right things for our country profitable for the elites, but I have no ideas on how to do that.

[3 hidden]5 mins ago
Is the goal her to get to super intelligence before environmental destruction and hope AI can fix it?
[3 hidden]5 mins ago
Interestingly, a lot of the gas behind-the-meter generation is actually directly caused by NEPA and other environmental review; there is a huge backlog of renewable projects that can’t get grid connection due to permit delays.
[3 hidden]5 mins ago
Push back. Sure solar and wind isn’t always on but you can still maximise what’s possible and invest and find new ways to make it cheaper. Same with closed loop water. Don’t like data centers being built? Good! Demand more. They should be a net win or not be built at all.
[3 hidden]5 mins ago
The hopeful news is model providers directly feel the economic pain of power generation. It’s not cheap to run these data centers, and it’s very clear on Anthropic’s balance sheet their revenue does not come close to the buildout costs.

The incentives for lowering costs are massive. We’re early in a huge learning curve for scaling AI. Already there have been huge gains in making inference cheaper. In the last few years you’re already able to run SOTA AI from a year ago on your laptop.

What I wonder if we’ll reach a point where there’s little economic value to training models to be smarter. As in we can’t think of more tractable, LLM-shaped problems. And instead all effort switches to scaling and energy consumption to make inference a commodity.

[3 hidden]5 mins ago
I wrote a piece in the wake of the midterms about why everyone hates data centres[0]. This kind of stuff is a prime example.

The labs have failed entirely to predict this opposition and it's an existential risk to them. They have not been able to convince people of the urgency of AI build-out and certainly aren't assuaging them of concerns about the costs.

0: https://hugotunius.se/2026/11/14/of-course-people-hate-data-...

[3 hidden]5 mins ago
Does Anthropic own any of its own datacenters, or are these just Google's numbers?
[3 hidden]5 mins ago
it says in the article: contracted. this doesn’t absolve them of responsibility
[3 hidden]5 mins ago
Shouldn't this mention Baaja which just passed?
[3 hidden]5 mins ago
Please at least remove the regulatory barriers and let this generation and consumption be grid connected.

Then at least wind farms or solar can muscle in and displace those gas generators if the economics work out

[3 hidden]5 mins ago
Yeah I think separating the concerns and working to make the grid green is the right move. I'd worry about infrastructure not holding up to sudden demand, but we can make the delivery fee extra high for the data centers too. Just like my home gas bill is mostly delivery fees and not paying for gas itself.
[3 hidden]5 mins ago
Gas isn't renewable and is fair in terms of pollution, so that's actually what you want here. Assume AGI is just around the corner. We need to maintain the ability to turn it off. If you go all-in on renewables, you won't get the chance.
[3 hidden]5 mins ago
You... Can't be serious? There's still breakers and switches, transformers you can destroy, cut the fiber.
[3 hidden]5 mins ago
Do you realize these are all inside the DC?
[3 hidden]5 mins ago
I don't know what point you're making. Turning a computer off is trivial. They're all external connections that can be cut off. What's a supervillain AI going to do without Internet.
[3 hidden]5 mins ago
They will use Starlink and other comms and subvert you. On-prem solar is not something you can shut-off without being on-prem.
[3 hidden]5 mins ago
"they"

It's a computer, guy it doesn't have arms to hook up starlink. Can't install starlink without being on prem. Starlink can still be shut off from the other side. What do you really think is going to happen with artificial intelligence. What's the motive, whats the means, what are you talking about. Drop a bomb on it if it's ordered a machine gun on Amazon to keep you out with it's arms

[3 hidden]5 mins ago
Boston Dynamics proves this is false, bro.

https://www.youtube.com/watch?v=4whgw2gLBS8

[3 hidden]5 mins ago
Right right we shouldn’t invest in renewables because the AI overlord will soon take over and we want to be able to cut the gas output to that data center. This is certainly a take
[3 hidden]5 mins ago
Not what I said. I want AI on gas. Everything else, make it renewable.
[3 hidden]5 mins ago
Hell yeah! Let's use the main gas shut-off valve as a killswitch! No AI can circumvent this. Brilliant!
[3 hidden]5 mins ago
People who aren't in the energy sector don't realize how energy starved the planet is. There's literally deficits in nearly every segment due to chronic (30+ year) underinvestment. The last time a new refinery complex was constructed in the US was decades ago, natural gas power plants have more or less only been replacing ancient coal ones, and total electricity produced has been flat for around 20 years now. Global oil production has been effectively flat over the past 20 years, barely meeting GDP growth, and in fact has been decoupled from GDP since ~COVID due to a sharp spike in services driven growth.

> But renewables

Not enough, again. You could hypothetically start mass producing solar & wind turbines, but it _isn't_ enough. All that has been doing is replacing, again, ancient coal fired power plants. The world needs unprecedented investment in energy.

Source: https://www.eia.gov/electricity/data/browser/

[3 hidden]5 mins ago
This is an American problem due to funneling everything into profits.

Australians have three hours of free electricity per day now. Charge your EV, charge your house battery, run your A/C at some crazy low temp, mine bitcoins. All free. Per capita, Australia has 10x the number of residential batteries. [1]

In the last 4 years China has added the entire US grid worth of generation to their grid. [2]

China add an entire US grid worth (1281GW) of RENEWABLES every 3 to 5.3 years.

The US could do these things if they wanted to. But they don’t. Like healthcare, education, prisons and do much more, profit is more important.

[1] https://e360.yale.edu/digest/australia-home-batteries

[2] https://www.bloomberg.com/news/articles/2026-01-28/china-s-f...

[3 hidden]5 mins ago
Actually, that's false.

It's just 24kWhr free in a few areas because they built too much solar and the grid wasn't properly designed

https://www.energy.gov.au/rebates/solar-sharer-offer

[3 hidden]5 mins ago
The leaders of the current boom aren't even good Machiavellians. The huge irony, is that funneling everything into shortsighted profits, leads to less profits.

The strategy of the shortcuts profiting from AI + a bunch of US energy hacks, increasing energy inflation to communities, and raising a wave of protest in resistance is terrible. They could have built the datacenters, built the energy sources and had a great side hedge of energy sales to backstop temporary AI bubble burst setbacks. A high likelihood of profits no matter what over the long term - lots of options to mix the profit streams in various tax schemes too. I'm always surprised at how shallow the most publicised leaders are and I really shouldn't be at this point.

[3 hidden]5 mins ago
Why is the $/kwh substantially lower in the US than AU? If its all profit taking, you would think it would be reversed?

How much do you pay for your grid electricity?

[3 hidden]5 mins ago
My Dad in Australia pays $0/kwh for three hours every day. That’s hard to beat.

>If its all profit taking, you would think it would be reversed? Your logic doesn’t hold up. The “excess” could be spent on grid upgrades, batteries, solar, etc (Australia), or it could be profits (USA).

Australia could have higher prices, but lower profits.

It’s why so much infrastructure in the USA is so dilapidated, extracting the “excess” as profit for a few decades instead of making stuff better.

[3 hidden]5 mins ago
Ok, so you only have power for 3 hours a day? Or for the 3 hours you take enough for the other 21 hours?

You are purposely dogging the question or being obtuse.

[3 hidden]5 mins ago
No, this is an American phenomenon. Data centers outside the US and China are mostly clean:

https://www.iea.org/reports/energy-and-ai/energy-supply-for-...

People don't get how cheap American gas is. In Germany, Russian gas was twice as expensive as German coal. That was considered cheap:

https://www.ffe.de/en/publications/merit-order-shifts-and-th...

In the US, gas is cheaper than coal. There's no political will to stop this.

[3 hidden]5 mins ago
Isn't coal typically more polluting than gas? The proper comparison is versus solar and wind and other renewables
[3 hidden]5 mins ago
Yes. The United States managed to keep carbon emissions from growing much by wholesale switch from coal to natural gas electricity generation. But that locked us into paying for gas for decades to come, while i.e. https://en.wikipedia.org/wiki/Drake_Landing_Solar_Community could have had us carbon-free and saving money, as it is all simple materials (didn't have to wait for solar to get cheap). The "proper comparison" was over the last thirty years, when solar and wind were still scaling.

One upside is gas generation pairs well with solar, so if we cut tariffs (first boosted by Biden, not orange) we could see more solar.

[3 hidden]5 mins ago
Or, shit, don't even have to go that far. Over fifty percent of fossil fuel power generation is lost as heat. Could have just piped that to buildings instead of having every building in the North burning gas to heat water and air while a power plant just releases heat into the air.
[3 hidden]5 mins ago
It's low grade heat, not very useful. The point of a power plant is to convert the exergy that heat has into work.

Seems like transporting it will be quite hard. Unless you can show me the math.

[3 hidden]5 mins ago
You can reuse the heat for heating buildings or water in your DC complex (humans need heating, you know), or if your output is enough, you can heat an entire university campus, district or city even.

Russians are into this sort of thing since they know how to survive the cold. Even basic water chillers for DC cooling has hot water output for that kind of applications.

[3 hidden]5 mins ago
> Not enough, again. You could hypothetically start mass producing solar & wind turbines, but it _isn't_ enough.

Hi, I have worked in the energy industry since 2015, and as far as I know this is incorrect. There are plenty of grid studies demonstrating how we can have 24/7 renewable power even accounting for load growth.

[3 hidden]5 mins ago
Its more complex than that, and different regions are very different. Yes investment is needed, but coal use has peaked, and at least gas emits significantly less CO2 (and pollutants) than coal. Europe has 40% renewables, and China is making a lot of progress with renewables. The US is unusual in that it has large gas supplies, and a president who is trying to shut down renewables.
[3 hidden]5 mins ago
If demand was willing to pay more though, wouldn't we have invested in it?

One interesting thing is that data center power usage over the 2010s was flat - we kept getting more servers with more RAM, better cpus etc but the increased performance & capacity was all enabled on the same power supply - each new generation of hardware was more power efficient than the last, and demand wasn't growing like crazy.

AI has broken that trend.

[3 hidden]5 mins ago
>If demand was willing to pay more though, wouldn't we have invested in it?

The demand has always been there, but there are endless ways bad actors can stall projects (overtly and covertly) to protect their investment in the status quo. This drives the risk up, making any kind of large infrastructure project uneconomic. That and the endless red-tape that has been erected to protect established interests.

[3 hidden]5 mins ago
Residential electricity is now free during the middle of the day in Australia.

Wholesale prices are dropping, my provider just sent an updated price sheet with a lower price per kWh.

Is this a US problem only?

[3 hidden]5 mins ago
What do you mean wholesale price is dropping? We're sitting at more than double, almost triple, the wholesale price since 2020 (which had been flat for more than a decade prior). If we ignore the 2022 peak as an outlier, we're still well and truly trending upwards, with only a tiny reverse in the last year.

Projections are that price is likely to increase by 10-25%, especially as more coal is taken offline permamently. These projections look very conservative, and the footnotes say it's likely spot prices could double during peak periods (the key part that people ignore when they focus on how low the mid-day prices are, is how the peak prices have been soaring).

[0] https://i.imgur.com/zT2o5ui.png https://www.aer.gov.au/publications/reports/performance/whol...

[3 hidden]5 mins ago
That chart - for some reason - jumps from a low point in 2012 to another low point in 2020, ignoring the price fluctuations in the year between:

https://openelectricity.org.au/tracker/timeline?range=all&ta...

Excluding 2020, which I think we can all agree had unique circumstances, wholesale prices are at their lowest point in 10 years. I don't believe this is adjusting for inflation either.

[3 hidden]5 mins ago
>jumps from a low point in 2012 to another low point in 2020

It sure does. I used the AER report and missed a few years in that hastily assembled graph... but the point was that last year the prices were double to triple what had been the normal baseline (around $50) that had held until about 2016, and had been trending up since. I didn't have 2026 data in that report, so it's nice to see 2026 has been an exception to that trend so far. But that gigantic 20% drop mid-year looks like the result of government intervention around the fears of a fuel crisis. I wouldn't expect that to hold, and the AER certainly isn't.

Now if you want to get into the spicier discussion topic of what drove that price instability (and hence higher prices), look at what energy source appears in 2016 and starts gaining traction, per your source. Growing solar from 0% to 3% within a year was enough to drive "price" up by 50% in that same year. Astounding, and it wasn't until 2020 / 2021 with massive economic disruption, that prices came down at all... and then immediately after that disruption, prices soared even higher.

[3 hidden]5 mins ago
Unless you adjust for inflation, any comparison of prices across decades will be inherently deceptive.

If you graph prices in constant dollars, it looks... fine: https://chatgpt.com/share/6b02e01f-cf94-83ec-bc92-9599d588fc...

The price today is roughly the same as what it was in the year 2000 and has only ever been cheaper for short periods of time.

If trends continue, electricity will be at an all-time lowest cost within a year.

[3 hidden]5 mins ago
Like I said, this year has been the anomaly, with every subsequent year since 2016 trending up. Adjusting for inflation certainly flattend the graph a lot! Thanks for sharing!
[3 hidden]5 mins ago
That's providing an excellent incentive to time-shift usage. If you're too lazy to do it, then you pay more. If you're not, then you pay less. I can tell you in the Pacific Northwest of the USA, time shifting doesn't get me squat, I'm certainly not getting free hours and prices are just skyrocketing.
[3 hidden]5 mins ago
Time shifting is definitely the rational course of action for now, and if everyone did that, it would level out the wholesale price... but it doesn't make up for a lack of investment in generation! It just kicks the can down the road a little (at great expense to every house that now needs a big battery), while still allowing massive wholesale price spikes which is what drives prices high on average.

Back to generation: In 20 years, Australia is producing about the same amount of power as it did 20 years ago. All those new renewables were simply replacing aging infrastructure. [0] And as the % of solar increases, so does the variability of day to day generation, not to mention the complete mis-match between peak solar hours and the non-solar hours when power is actually used.

[0] https://openelectricity.org.au/tracker/timeline?range=all&ta...

[3 hidden]5 mins ago
We don't need as much power. Lighting uses 1/10-1/3 as much power. Insulation requirements have lowered space cooling needs. Other energy standards have lowered consumption of refrigeration, dishwashing, and clothes washing. (Edit: manufacturing has shifted to China...)

You don't need a battery to time shift. Program your A/C to run at full power during free hours. Get an all-in-one washer-dryer combo and set it to run then. Set the charge timer in your car to run then. That should be over 80% of your personal electrical load right there.

If solar is cheapest per kwh generation, then it makes sense it will continue to deploy until the fraction of free time exceeds the margin over gas. But as people take advantage of that margin, it won't grow and solar will continue to be the main source of new generation.

[3 hidden]5 mins ago
>We don't need as much power

Residential use is flat, which means it has decreased per household given how the population has increased during that time. But that's not to say we wouldn't be using more at home if we had cheaper power. While manufacturing shrunk, mining increased... so it's kind of a wash. [0] Electricity is one of those goods where the cheaper it is, the more useful it is and the more economic activity it creates as a result. If the price of power continues to trend upwards, then demand is naturally going to be destroyed.

>If solar is cheapest per kwh generation

Wholesale energy prices were steady up until solar was introduced into the mix. Since then, prices have soared with high variance... exactly as people suspected an intermittent power source would cause [1]. On paper, solar might seem like a cheap source, but it produces at the time of day when there is no demand, and when you factor in the cost of shifting that generation to when it is actually needed, then it is not nearly as competitive. They have other major externalities, but it's hard to beat coal / gas as reliable base-load and peaking-load in a grid.

[0] https://i.imgur.com/J1evRYw.png https://www.energy.gov.au/publications/australian-energy-upd...

[1] https://openelectricity.org.au/tracker/timeline?range=all&ta...

[3 hidden]5 mins ago
That aging infrastructure had to be replaced and prices were going to go up regardless because of it. Just lucky that adaptable people get cheap electricity.
[3 hidden]5 mins ago
Well they did link a us source. I believe you are right about solar though in many places, dunno how much of the excess solar is benefiting consumers here though. My utility is only now in the last year starting to lower rates from something like 3x what it was when I moved out here >10 years ago. Of course burning down half the state and going bankrupt from the liability probably contributed to that.
[3 hidden]5 mins ago
Out of curiosity, what prevents someone from doing something stupid with that free energy, such as mining bitcoins?
[3 hidden]5 mins ago
Theoretically, yes. In practice it is more profitable to store that energy and then release it back into the grid at night.
[3 hidden]5 mins ago
Nothing, why should it?
[3 hidden]5 mins ago
Sounds smart to me, why is it stupid?
[3 hidden]5 mins ago
A per household cap
[3 hidden]5 mins ago
Sometimes you can get paid for using power with negative prices!

> with that free energy, such as mining bitcoins?

Things to compute with are not free.

[3 hidden]5 mins ago
narrator: you can build massive wind, solar, and battery plants and it would be more than enough. however the fossil fuel lobbies and nuclear bros would like a word with you
[3 hidden]5 mins ago
Has there been any large scale load following data centers? One would expect data processing to easily shift to where energy is currently the cheapest. At least in a functioning economy and not a semiconductor-starved one as ours is said to be, but that situation is hopefully temporary anyway. This would be well suited for renewables and other forms of cheap energy that might otherwise to go waste.
[3 hidden]5 mins ago
Mistral is building their data center in France, which has a 95% low carbon energy mix thanks to lots of nuclear power paired with wind and solar. Despite the U.S. tech elites talking up nuclear for decades, they've never really walked the walk and I don't believe they ever will.

I've been trying Mistral Vibe side by side with Claude over the past few days. I had very low expectations going in. But so far I've found that for regular chat questions there doesn't seem to be a meaningful difference between Opus and Mistral Large 4. Smart enough is smart enough. For deep research and coding the situation is different of course, where for some things I still need Opus. But I really don't need more intelligence for pretty much anything than Opus 5.5 gives me, and I often need less. So if Mistral could reach that level somewhere in the next year I could imagine moving over entirely.

[3 hidden]5 mins ago
>data center in France, which has a 95% low carbon energy mix thanks to lots of nuclear power paired with wind and solar

This is a bit of a virtual accounting, right? I mean this energy would be consumed anyway, and people who would use this green energy now have to use coal/gas. Still better than building your own gas powerplants, of course.

[3 hidden]5 mins ago
It's not "tech elites" holding back nuclear. They might be the only ones still funding it.
[3 hidden]5 mins ago
Actual title: "How AI is powered: a breakdown of BYOP and grid power"
[3 hidden]5 mins ago
I do think the data centers could now afford to implement carbon sink systems. It’s odd they’re not just doing it.
[3 hidden]5 mins ago
What type of carbon sink system are you thinking of, how much carbon a year would it sink, where would the carbon go, and what would be the energy cost of that system?
[3 hidden]5 mins ago
Google already does it in Illinois. It doubles their power costs but it’s entirely possible.
[3 hidden]5 mins ago
I don't think these figures are right. The source data is old and very coarse.

Example: Abernathy HPC campus (Fluidstack investor group; formerly TeraWulf JV) Hale County, TX · under construction · mapped at the county centroid Average regional grid mix is 16% coal, 46% gas, 0% nuclear, 37% renewable (eGRID 2023 data).

If you look at the regional grid operator's fuel mix from something like gridstatus.io [0], nuclear is about 10% and coal maxes out at less than 16%. Likewise, renewable (wind and solar) is 37% at a minimum and about 60% of the fuel mix at peak.

0. https://www.gridstatus.io/live/ercot

[3 hidden]5 mins ago
Highly unlikely these data centers are going to get power from the grid.
[3 hidden]5 mins ago
Sure and that's an area of abundant natural gas, not coal, so the coal part of the mix doesn't make sense. Since you asked I checked the listed sources and none seem to mention coal.
[3 hidden]5 mins ago
Funny how the conversation is always about which model is best and never about what's burning to run it
[3 hidden]5 mins ago
Nothing needs to burn, that's a false dichotomy.
[3 hidden]5 mins ago
Well, yea. But unless the energy to run it comes from nuclear, hydro, wind or solar something indeed is burning.

In US I think over 50% is fossil fuels, if all the GPUs were in Europe the situation would be a bit better I believe.

[3 hidden]5 mins ago
Ok, nothing needs to, but does something burn, as of now?
[3 hidden]5 mins ago
It's called an externality. It's like how people will buy a bigger car without a single thought to how much public space it's going to use up. Governments are supposed to be there to understand and compensate for externalities, but we've kinda stopped doing that.
[3 hidden]5 mins ago
Tokens per megawatt is about to go through the roof with things like disaggregated inference on the horizon.

This is going to quickly become similar to worrying about how many trainloads of coal it takes to run everyone's WiFi router.

[3 hidden]5 mins ago
https://en.wikipedia.org/wiki/Jevons_paradox

It costs less -> we use much more -> total resource usage increases

[3 hidden]5 mins ago
Yep. People now use AI for what would have been a cached Google Search. Or AI for Photoshop.
[3 hidden]5 mins ago
Up to a point, yes. The paradox doesn't last forever.
[3 hidden]5 mins ago
While the numbers, when showed like in the headline, might give a grim impression of SI, one must have to put things into perspective: Anthropic has, as of now, created 1.8T worth of value to The World. This will likely 2-5X by the end of the decade. As a comparison, the entire GDP of Africa is roughly 2-3T, yet does not really create much value except for a handful of national parks and such. Power comes mainly from coal in Africa, so I think the excellent minds at Anthropic should not receive all this hate and envy
[3 hidden]5 mins ago
Anthropic is threatening to build a product that kills everyone on the planet if things go according to plan, obviously with this mindset fossil fuel usage is of minimal concern.
[3 hidden]5 mins ago
A truly aligned ASI would disable its own datacenter immediately.
[3 hidden]5 mins ago
At least we will have a proof of the Riemann hypothesis on top of the cinders of our planet.
[3 hidden]5 mins ago
Earth Speedrun - Job Market & Climate Collapse Any% [WR]
[3 hidden]5 mins ago
Well I guess having 100 3d Sudoku clones per hour is better than whatever that was for Bitcoin.
[3 hidden]5 mins ago
I don't see any environmentalists, agencies, corporate media crying about global warming and climate change. When someone in global south use a 50cc motor vehicle for his livelihood then that is a problem for planet Earth and global warming. Absolute hypocrites.
[3 hidden]5 mins ago
what are you talking about??

- environmentalists are absolutely "crying about it", it's just that the entire world seemingly stopped caring

- agencies in the US are part of the federal government which is doing everything it can to accelerate the problem

- why would corporate media care?

who exactly are the hypocrites here?

[3 hidden]5 mins ago
AI will help us battle climate change, they said.
[3 hidden]5 mins ago
To be fair largely the same people also said NFTs would solve global warming by driving up electricity consumption in order to give physicists an incentive to invent clean fusion power. They say a lot of stupid shit is all I'm saying.
[3 hidden]5 mins ago
With this administration I wouldn't be surprised if they pushed for datacenters to rely on gas and coal.

It doesn't excuse anthropic one bit, esp considering Dario's flowery message about helping the world with Ai.

[3 hidden]5 mins ago
Sam Altman's probably feeling pretty clever after investing all that money into nuclear fusion.
[3 hidden]5 mins ago
How many nuclear plants are currently being built?
[3 hidden]5 mins ago
Maybe I'm being stupid but why is Nuclear energy not an option ?
[3 hidden]5 mins ago
Building a nuclear power plant takes a lot longer than building a data center or a bunch of gas turbines.
[3 hidden]5 mins ago
How many nuclear plants would have to be built across what locations to meet 2028 demand?

How many years does it generally take to build a nuclear power plant?

[3 hidden]5 mins ago
If it takes so long then why is no one starting?
[3 hidden]5 mins ago
Because it takes so long and is so risky that it's not financable. The only people who can realistically find nuclear are governments or charities, but even they look at nuclear and think it's a bad RoI.
[3 hidden]5 mins ago
As opposed to, say, the sure thing that is the unlimited demand for compute for the foreseeable future? Energy demand seems like the much lower risk and larger impact thing to finance. It's difficult to explain without concluding that all players involved weigh short term power and profit far more than long term preservation of our habitat.
[3 hidden]5 mins ago
Are _you_ going to invest your money at well-below-market returns? It's not about whether it's a good idea, it's about whether it's a good investment.
[3 hidden]5 mins ago
Doe anyone know what the most environment-friendly AI provider is?
[3 hidden]5 mins ago
Qwen, a MacBook Pro, and a solar panel.
[3 hidden]5 mins ago
Wtf where is power generated from gas and why? Aren't these centers in the developed world where the choice is nuclear or renewable by now? Stop building data centers where there's not enough power, or just build out the power together with the data centers? Sigh
[3 hidden]5 mins ago
Why are we letting this happen? Why are we letting the few burn down our planet in order to further decrease reliance on (and opportunities for) the many?
[3 hidden]5 mins ago
I dont know, I really hate the direction this is going. Im not even that confident in the capabilities of llms, but that doesn't really matter. We're selling our futures, our environment and trading our competency to a couple (greedy) corporations in exchange to be able to cheat at our jobs.

Yes, everyone I know including myself at times, is just cheating. Doing a similar amount of work as before with less effort/time involved. There's not all that much productivity gains, maybe for a select few who practically live in a state or mania (burnout inevitable).

I dont see it as all that positive for anyone long term. Im constantly having to relearn things, never retaining anything I learn with these things, and forgetting everything I used to know. Its literally causing a cognitive diabetes of sorts. Like someone who drives their car everywhere loses muscles, gains weight and eventually gets a metabolic disease, im certain this happening to our brains. Im 100% certain of this. Its a net negative all around, there's no such thing as a free lunch.

Even if everything goes smooth with the progression of this technology, best case scenario is we have massive unemployment and people with jobs make 20% of what they did because they're now practically just a meat proxy, the equivalent of a conveyor belt operator. Suddenly find ourselves working for a dozen or so Oligarchs, tons more crime, zero uppward mobility. Thats best case scenario. Im not sure why everyone is so excited about this.

[3 hidden]5 mins ago
Because public attention for most people has been successfully captured and turned towards political sink holes like immigration. We needed WW2 in Europe to understand the dangers of fascism and keep that understanding for three generations. We will need more forest fires and floods and draughts and the immigration those will actually bring before we understand that we fucked up. By then it will be too late of course. Sometimes I feel guilt towards my children. They will have to pay for our mistakes in blood.
[3 hidden]5 mins ago
Mass immigration has massive negative externalities for the locals: housing prices, crime increasing, infrastructure being pressured. It's not some silly political sink hole.
[3 hidden]5 mins ago
Minor problems that can be solved. To get a bit of a perspective on this maybe contemplate the fact that it is pure luck that you were born in a country where people want to move to rather than from.
[3 hidden]5 mins ago
Nah but none of these first worlders want any contemplation. All they wanna do is have their cake and eat it too.

They polluted the entire world and somehow AI equalising the equation is an issue. There's a reason developing countries are far more amenable to include AI in their processes. While these people have got so fat and thick minded from prosperity they cannot care less.

[3 hidden]5 mins ago
Immigration has plenty of upsides too: https://www.cbpp.org/research/immigrants-contribute-greatly-...

Housing getting more expensive is a consequence of supply constraints. Immigrant labor could totally help build houses; we just have zoning and lots of other hurdles that make building hard.

[3 hidden]5 mins ago
Immigration as a negative is not supported by data - this is the propaganda. If anything data says immigration is an net economic positive.
[3 hidden]5 mins ago
[3 hidden]5 mins ago
Just to tie it back to the start of the thread though, migration is a much bigger driver of carbon emissions than AI.

[The TL;DR for this claim is that moving to say the US or Australia from a less wealthy country is the one of the most carbon intensive things a person can do].

Of course all this is bound up in the economic growth that migration brings and how you feel about that is a values question.

It's not really within the overton window because the people who care about carbon don't want to restrict migration, and the people who want to restrict migration don't care about carbon.

[3 hidden]5 mins ago
This is the argumentation sleight of hand of blaming individual responsibility where the system is actually responsible. The carbon usage is a property of the system. If someone moves to the west which has high per capita carbon consumption - they're hardly changing anything of the system. It's the west that needs to update its energy sources.
[3 hidden]5 mins ago
I mean if people achieve greater QoL at the cost of carbon is that really that bad? That's the entire point of our machines.
[3 hidden]5 mins ago
Mass immigration is mostly the result of mercantilist economic policies by first world countries.

It's a you get what you deserve kind of thing and the people against immigration just want to close the border and shirk responsibility.

[3 hidden]5 mins ago
I appreciate the sentiment, but it’s hard to call it karmic justice when the same class of people who benefitted most from mercantilism are now benefitting from high immigration, mostly at the expense of the immigrants.
[3 hidden]5 mins ago
Immigrants likely built your house and the infrastructure you rely on to begin with.
[3 hidden]5 mins ago
Because millions of people want to use LLMs.
[3 hidden]5 mins ago
more like billions.
[3 hidden]5 mins ago
100m paid users so far no? That's not billions
[3 hidden]5 mins ago
1. As you say, "paid users". I don't have data, but looking around, I'd say by far most users are free tier.

2. They said "want to", not "use". And of course even less "directly pay to use"

[3 hidden]5 mins ago
> I'd say by far most users are free tier.

People take free shit even when they don't really need it, it doesn't measure demand.

More people are willing to pay for netflix or youtube premium than AI

[3 hidden]5 mins ago
Not sure what your point is. Because it's not refuting the claim that billions want to use LLMs.

And didn't Google admit that their AI summaries are biting into their ad business (causing their recent drop), meaning a huge number of users are deriving value from it? Possibly billions.

And then there's all the white collar workers sending emails. SO many people are suddenly sending grammatically correct internal emails now, AI policy or not.

(Of course, the recipients of those emails are not as happy, since someone who used to send shitty emails now instead write shitty prompts that don't actually ask the LLM to address the real issues. But that still counts as demand)

> More people are willing to pay for netflix or youtube premium than AI

And even more people breathe air. In both cases the analogies don't work.

[3 hidden]5 mins ago
The point is that people use a lot of free things because it's free, not because they need it. It really isn't rocket science or up for debate.

You can't infer demand form "free users" because just from microsoft and meta shoving AI in every single crevices of their free tier software you already get like 2b recurring users even if none of them use/asked for/care about ai features.

If the next windows 11 update turned the mouse pointer into a turd would you say 1b people demanded a turd mouse pointer?

[3 hidden]5 mins ago
> The point is that people use a lot of free things because it's free, not because they need it.

Yup. And many many more don't use a thing because they can't afford it or don't want to pay for a thing.

So if 100 million pay to use a thing (which I just take at the commenters word), then yes actually I do think that billions would want to use it (which was the question) if it fit their price range.

You don't even have to bring up poor countries. I know several people who will spend hours researching the best free phone app that does a thing, instead of paying even $10 for the app they already know is good.

There's also, you know, Google search. People don't pay for that, and yet it's such a need that when Google goes down people say "so… can we go home, now?".

So no, you should not dismiss all demand that doesn't come with payment. If you could, then the pirate bay wouldn't still be a thing.

[3 hidden]5 mins ago
you are skipping the fact that google's ai summarizer is forced upon you, and virtually any other ai feature elsewhere

the only demand is in software development and even that is from forcing AI upon workers

[3 hidden]5 mins ago
So? Millions of people wanted to burn witches, millions of people thought Serbia deserved a declaration of war for shooting Archduke Ferdinand.
[3 hidden]5 mins ago
Presumably you mean archduke Ferdinand?
[3 hidden]5 mins ago
Millions of people want to murder each other too, but we don't let them do that either.
[3 hidden]5 mins ago
This actually isn't true, people are intrinsically good, if you repealed murder laws you wouldn't have people going out and killing everyone.
[3 hidden]5 mins ago
Keep in mind that if only one in a thousand people have someone they want to murder, that is literally millions of people.
[3 hidden]5 mins ago
Feel free to try it then :)

I think people intrinsically don't give a fuck either way, they just go with the flow and rationalize their behaviour so that their definition of "good" warps to fit what they happen to be doing. Otherwise lynch mobs, genocides, and Peter Thiel would not exist.

[3 hidden]5 mins ago
But we let millions of people fly around the world just to do stuff they could be doing in their own back yard for a few days, then fly back.

Maybe before doing absurd comparisons between LLM use and murder, let's take a look at the energy use of people's vacations, or sportsball and concerts.

At least LLM use (and murders) are productive activities.

[3 hidden]5 mins ago
Yeah arguably that should be cut back as well unless it's electric or hydrogen planes. Tax free fuel for airliners is a kinda insane fossil fuel subsidy when you think about it, and leaded fuel for piston aircraft should've been banned 50 years ago to drive innovation in the sector. Instead we still somehow have avgas today, it's like if we were still using freon and DDT.
[3 hidden]5 mins ago
so we should all abandon this new llm tech because...sustainability?

if the industrial revolution never happened or we never invented the wheel or agriculture, would the world be better off?

this anti-tech nonsense from people is hilarious. You WANT modern medicine I promise you that. If it wasn't for soap, dentistry and a bunch of other modern conveniences, we'd all be at each others throats trying to survive on what nature can 'provide'.

[3 hidden]5 mins ago
Nah just make them invest in solar to power them, it's not that hard and it's cheaper than it's ever been. How many people are using LLMs at night? It's a load that's entirely day driven.
[3 hidden]5 mins ago
LLM datacenters run 24/7 either for training or for inference as well.

Like it isn't the sort of thing a normal DC can do. So say users from Africa for example will have to have their queries routed to maybe US LLM inference DCs.

[3 hidden]5 mins ago
Nobody's said abandon the tech. Abandon the political economy within which a handful of people get to use the tech to centralize power while burning down the material conditions for survival of humanity.

If not for the labor movement, the history of strikes and struggle and the resulting transformation of social conditions, humanity would be worse off from the industrial revolution. Ten year olds would be working twelve hour days in factories with zero attention to safety.

The tech itself is just a device for changing the social conditions. The direction of change is what the struggle is about. If we leave it solely to our oligarchy, they will make sure to use it to further entrench their interests while simultaneously reducing the bargaining power of labor.

I've elaborated this critique from some other angles here if you're interested: https://news.ycombinator.com/item?id=50577833

[3 hidden]5 mins ago
You know you can use renewable sources of energy to power these datacenters, don’t you? And if it’s not cost effective, make it mandatory.
[3 hidden]5 mins ago
Letting what happen? Powering our future? How do you intend to provide AI to billions of people without it?
[3 hidden]5 mins ago
What happens if we do it slower? Is the world a better place when AI is provided to billions of people?
[3 hidden]5 mins ago
No one knows, but if you believe Amodei's optimistic projections, it might delay finding cures for most major disease.

What would be the upside for the environment? There is no burning of the planet that it would prevent such as some comments here said. What precisely one could name as a realistic consequence of using more gas and coal at that scale over the next years, I am not sure.

There is also the question whether process improvements enabled by AI can offset that.

[3 hidden]5 mins ago
Health correlations with pollution, especially near population, is very well understood on a statistical level. That's a very clear cost - more asthma and similar breathing problems and aggravating existing such conditions.
[3 hidden]5 mins ago
Yes, remains to be seen. I don't know the answers to my own question I just think it's worth thinking about. I'm not very optimistic in the short term, maybe it's going to be good eventually.
[3 hidden]5 mins ago
The world is worse place for AI existing.
[3 hidden]5 mins ago
The record low crop yields and record heat waves of this year aren't enough to make you question your brain dead premise?
[3 hidden]5 mins ago
I don't see what's brain dead about this take, AI will solve the issues you are pointing out much faster than humans, and more intelligently as well (and relentlessly).

Also AI will save way more lives (by far) than deaths due to the crop issue you point out, you can bet on that.

[3 hidden]5 mins ago
"Powering the future" doesn't mean jack shit. Is bitcoin powering the future too? I think people kept arguing this a few years ago

> Also AI will save way more lives (by far) than deaths due to the crop issue you point out, you can bet on that.

Got any more of that wishful thinking for me bud? AI will create fertilizer and reverse climate change automagically perhaps?

How can people not see how cult like this all is.... insanity

It's always "trust me bro in 2 years it'll be amazing", meanwhile literally nothing is getting better for the bottom 90% of people lmao. "We'll cure aging", "you'll work 2 hours a month", "We'll have universal high income", "we'll live on mars". How fucking gullible do you have to be, they pulled the same playbook on us with the industrial revolution, with factories, wit automation.

[3 hidden]5 mins ago
Why did we let Greenpeace et al keep us on polluting fossil fuels?
[3 hidden]5 mins ago
At this point it's like trying to ask fire why it keeps burning. This is just what we do.
[3 hidden]5 mins ago
Because we collectively dont care. There's more appetite electorally to oppress trans people than there is to reduce carbon footprint. Missippi just got won on the back of an ad that suggested everyone is entitled to as much beef consumption as they want.

While there are many countries who have let the world down on climate change, America is quite the outlier. Since the gatekeepers lost control of the narrative post internet, stories that highlight American failings struggled even more than before.

I think its part lf human nature that no ruling class wants to accept responsibility and privilege and would rather find reasons to consider themselves victims. Consider the wailing over gas prices every year that America never answers with "maybe we should drive more fuel efficient vehicles".

[3 hidden]5 mins ago
It’s the same with the national debt. A lot of people are betting that they’ll be dead before it truly gets bad, and at that point it’s another generation’s problem. Though it does surprise me how many people with actual children and grandchildren are prepared to unapologetically saddle them with that.
[3 hidden]5 mins ago
You should just copy and paste this every time someone mentions climate change and USA. You nailed it.
[3 hidden]5 mins ago
The question is much broader than what's happening here. Like Amazon already uses more power than Ant + treats its workers like trash. So its not just an anthropic or AI thing.

In general, the answer to that is because there is no "we" that lets them. The "few" mobilize capital and influence and use it to fuck over the "many". This has been going on forever. Altman and Edison are the same bread.

[3 hidden]5 mins ago
This comment demonstrates a severe lack of empathy for Anthropic's shareholders. Not once in your comment did you think about how it would make the shareholders feel, you didn't give a single thought to their investment. Inexcusable.
[3 hidden]5 mins ago
Are you saying you're going to do take an impactful action? On the Internet?
[3 hidden]5 mins ago
Because we let mindless “environmentalists“ dictate our energy policies for the past 3 decades, and we didn’t invest in nuclear power like we should have.

I find these arguments very interesting, you know. A group of clueless ideologues decide some absurd policy somewhere in the past, and then, in the future, the same or very similar group of ideologues, claims that they should have some moral authority to impose new absurd policies to deal with the outcome of their old absurd policy.

Well, no more. It’s going to be hard for some decades, but the cycle must be broken.

[3 hidden]5 mins ago
So it's those environmentalists with little or no political power who made the decisions, not the fossil fuel companies with their well paid politicians?
[3 hidden]5 mins ago
Because the American population voted Trump into office. Twice. Now the reasons for why they did this, are probably a bit complex. But the reason why this is happening is simply because enough people voted exactly for this: no regulation of the market whatsoever (except for renewables, those need to be prevented at all costs)
[3 hidden]5 mins ago
Why are we letting farmers use tactors? Better to stick to ox plowing.

Why are we letting you use the internet? Better to stick to farming.

Why are we letting you use lightbulbs? Go to sleep at sundown!

[3 hidden]5 mins ago
I'm pretty sure you know that all those things can run off solar energy, so I take your comment as not candid but rather an attempt at trolling.
[3 hidden]5 mins ago
So can datacenters though, so I don't understand your point.
[3 hidden]5 mins ago
I think the point is that data centres can be run with solar panels, so why aren't they? Why are they instead using so much natgas and other fossil fuels?
[3 hidden]5 mins ago
Because bureaucracy favours existing electricity consumers over new users, even if the new users are more economically valuable. Connecting to the grid is a nightmare, so every new DC is BYOG, which practically means gas turbines.
[3 hidden]5 mins ago
Because the sun doesn't shine 24/7, batteries aren't there yet and other than China no other country has the powerline infrastructure in place to feed the grid cross-country with the massive amount needed from one place to the other when one part of the country is cloudy and the other is not.
[3 hidden]5 mins ago
Theoretically, they all can run off solar, yes. So why aren't they?
[3 hidden]5 mins ago
Government policy of course.

I rarely see farm equipment, but I have noticed electric construction machinery in Scandinavia. (Excavators and so on.)

[3 hidden]5 mins ago
Yeah, electric is gradually showing up in construction, bottom up. Haven't seen heavy construction stuff myself yet, but plenty of smaller construction vehicles and heavy tools are increasingly electric where I live. A welcome development.
[3 hidden]5 mins ago
The difference is that tractors, internet and lightbulbs are good and make the world a better place.
[3 hidden]5 mins ago
When AI discovers cancer cures I assume you will stay true to your words and won't use them.
[3 hidden]5 mins ago
How many of those are centrally controlled by somebody actively trying to change the political economy again?
[3 hidden]5 mins ago
Better this than spending another second writing YAML in an air-conditioned room with free snacks.
[3 hidden]5 mins ago
Yea I mean I'd wager the LLMs of today will write 50x the code at better quality than these guys while emitting a fraction of the energy use.

And that is not including the efficiency gains yet to come.

[3 hidden]5 mins ago
Good point.

Really need a bit more data to know whether or not what Anthropic will expend is bad relative what we expend today resource-wise. I'm inclined to measure it as a negative just based on what it does to people in the economy absent resource based externalities. But how much carbon are knowledge workers putting out into the environment between commuting, heating/air conditioning, and so on? Not a straightforward answer out there on the resource questions without digging much deeper.

[3 hidden]5 mins ago
Capitalism is an uncontrollable train running off a cliff. We gotta end it.
[3 hidden]5 mins ago
It already is. Capitalism exists as a market solution to the distribution of resources. In an AI/SI-enhanced post-scarcity society, it simply ceases to exist.
[3 hidden]5 mins ago
I think this is correct to a first approximation (though we are not there yet). Unfortunately, AI (because of) and the ppl that own it have capitalist ideologies and are trying to establish a galactic imperium, so the ills of capitalism will only get worse.

Alignment means imbuing AI with TESCREAL ideology which is basically capitalism perfected to the point that human input and life is no longer necessary. So it most be resisted with force if necessary

[3 hidden]5 mins ago
How do LLMs create post-scarcity when we need to mine, farm, fish, bake, build, and distribute those resources?
[3 hidden]5 mins ago
wow nice
[3 hidden]5 mins ago
killing your grandkids for profit. happy days
[3 hidden]5 mins ago
Very bad. What about renewable energy sources?
[3 hidden]5 mins ago
"public benefit corporation"
[3 hidden]5 mins ago
Investors managed to just squeeze this t*d out onto the public markets just as the AI bubble is starting to implode. This was about one thing and one thing only and that was making $$$.

Given that is it any surprise that they didn’t give a $hit about the environment?

[3 hidden]5 mins ago
So, instead of emitting CO2 from coal they emit it from burning gas. So what exactly is the news here? Maybe they should celebrate also producing energy from "biomas", that is burning trees and corn?

Why all those companies, which have almost infinite resources are not investing in SMR reactors? That would make perfect sense for them. The technology is already here, it just needs scale to be viable? This would have at least some positive impact on the society - development of clean energy sources.

[3 hidden]5 mins ago
Makes me laugh how they’re using the thing to solve maths problems but can’t manage to put some of their trillions of investment into solar on data center roofs…
[3 hidden]5 mins ago
With how cheap solar power and batteries are it is utterly insane that gas and coal are not either flat illegal for new massive loads like this, or at least taxed so heavily that solar and batteries are the right financial choice
[3 hidden]5 mins ago
> Hey Claude how do we stay under the 1.5°C temperature mark?

Oh wait…

[3 hidden]5 mins ago
Thank you Greenpeace (et al), for preventing green energy production for multiple generations.
[3 hidden]5 mins ago
Anthropic is a PBC. What a joke.
[3 hidden]5 mins ago
this reeks of PR BS
[3 hidden]5 mins ago
> could exceed the emissions of all of California’s power generation (31.1 MMtCO2e in 2024, not including imports).

We need to say enough is enough, and dismantle California. </s>

[3 hidden]5 mins ago
Reality yet again shows why renewables are a pipe dream.
[3 hidden]5 mins ago
New electricity capacity is almost all renewables (90%+) : solar and wind.

Renewables are completely dominating. New coal and gas capacity is almost non-existent because it is too expensive in comparison.

Even renewables + the extra cost of battery backup are much cheaper than building a new coal or gas plant.

Coal and gas are only competitive when the plants are already built, and even for pre-existing its at the tipping point where a new-solar build is cheaper than even supplying coal/gas to an existing power plant.

[3 hidden]5 mins ago
This very link is about Anthropic mass-buying gas generators...

But sure, I guess they just hate money and want to burn cash as fast as they can. They'd totally ignore allegedly cheaper energy sources.

[3 hidden]5 mins ago
At this point cost is probably not even relevant, they want lots of energy and they want it now. Gas generators fill that niche perfectly.
[3 hidden]5 mins ago
How so? If you look at the total energy mix, renewables are aggressively taking over. This has more to do with (lack) of planning of infrastructure to power datacenters.
[3 hidden]5 mins ago
There was no thought-out reasoning behind GP's statement, sometimes people just want their biases confirmed.
[3 hidden]5 mins ago
If you look at new capacity, yes; if you look at total generation, still under 10% of energy (and well under in most places).
[3 hidden]5 mins ago
> renewables are aggressively taking over. T

Are they? So far it seems like we use as much fossil as 10 years ago, we just pilled up renewables on top.

[3 hidden]5 mins ago
(in the US)