This reads as less than useful research to me. E.g.:
"The analysis does not account for the purchase of market-based instruments that are meant to represent investments in new renewable energy in the US and that tech companies buy to offset the pollution from their electricity consumption. ... 'Unlike carbon emissions, the health impacts caused by a data centre in one region cannot be offset by cleaner air elsewhere,' said Shaolei Ren, associate professor at UC Riverside."
Why not? Does clean air elsewhere not matter for the individuals elsewhere? Where is that "elsewhere?"
My admittedly ungenerous interpretation: factoring this context in would be really hard and would decrease the headline factor of the findings, so... publish without.
Some reports say carbon offsets suffer from widespread fraud [1] - if you emit pollution, pay someone to 'protect the rainforest' and they 'protect' some rainforest that wasn't at risk anyway? Your emissions haven't actually been offset.
Likewise if you pay someone to plant a tree that'll reabsorb the carbon you've emitted over the course of 40 years, and they cut it down and burn it after 15 years - the carbon's still in the atmosphere.
So putting your data centre right next to a hydroelectric or nuclear power plant is the gold standard.
[1] https://www.theguardian.com/environment/2023/jan/18/revealed...
> Why not?
If your house had a carbon monoxide leak but your neighbor’s house did not, would you consider your carbon monoxide leak to be “offset” by that fact?
If a mitigating action had decreased CO in my neighbor's house by at least as much as it increased in mine, then potentially yes, a neutral researcher should consider that as an offset at the population health level.
Not sure why we need an analogy, though. How about sticking to standard air pollution, which has direct impact of its own? CO poisoning seems a bit extreme as a comparison.
No, it's not that extreme. If the offset comes from a zone where air quality is already "good enough", no extra lives are saved, so it's still net negative.
Sounds to me like the real criticism is dirty electricity generation.
> Sounds to me like the real criticism is dirty electricity generation.
Not when that "dirty electricity generation" was planned to be shut down before the data centers arrived: https://www.theregister.com/2024/10/14/ai_datacenters_coal/, https://www.theguardian.com/technology/2022/feb/18/bitcoin-m..., https://www.theregister.com/2024/05/31/datacenter_power_crun..., https://www.theregister.com/2024/10/08/eric_schmidt_speech/.
Everyone needs a pollution tax, hardly anyone wants one
A lot of modern electricity generation is to power data centres and bitcoin mining.
Quick googling found a stat that 1.6% of the whole world's electricity supply going on computing and also https://www.visualcapitalist.com/charted-big-tech-uses-more-... etc.
And we need more! On day 2 of his presidency Trump pledged to fast-track new power generation to drive data centres in the US. https://www.axios.com/2025/01/23/trump-ai-power-plants-data-...
Drill baby drill! Using electricity on compute is driving future dirty power generation.
> Quick googling found a stat that 1.6% of the whole world's electricity supply going on computing
That would imply that it isn't the major issue. Electricity generation as a whole is less than half of US energy consumption, US electricity generation is 35% nuclear or renewable, and computing is using 1.6% of electricity. This is a rounding error compared to e.g. transportation at 37% of all energy consumption, or HVAC at another double-digit percentage.
> And we need more!
New generation capacity in the US is disproportionately renewables because renewables will remain cheaper until they're closer to half of all generation and then have to really deal with the storage issues. Willingness to approve fossil fuel plants doesn't mean anybody really wants them. The large tech companies are buying nuclear:
https://www.nytimes.com/2024/10/16/business/energy-environme...
Fuck future generations, fuck our kids... often touted by folks without any or at least somehow normally functioning family.
I think we are back in 80s with current orange man defining direction like fashion, 'greed is good' at any cost mantra is certainly back
Greed is why people are going to spend CapEx installing solar to reduce OpEx buying coal.
Greed pushes people out of the status quo.
Greed is why fusion reactors will be reduced in size, cost, and complexity.
I’m confused by the article mentioning West Virginia as a prime location for data centers along side Ohio - I am not aware of a large concentration of data centers in WV? (Unlike VA, of course)
https://westvirginiawatch.com/2025/01/16/west-virginia-isnt-...
In broad strokes, take AWS for example with us-east-1 being in Virginia and us-east-2 being in Ohio, respectively, and wondering how to fuel future East Coast expansion, West Virginia would offer similar latencies in a deep-red state (meaning, deals will happen when palms get greased) with a powerful energy lobby (that would appreciate additional energy demand). Being a poor state, the price of labor will be cheap as well, both to build the data centers and to maintain them afterwards.
Maybe the author thought West Virginia was the western part of Virginia? I know it sounds preposterous, but I think it’s commensurate with the quality of the headline and article.
Ugh, this is horrible writing.
For instance, "Data centres cause pollution through high eletricity use" - high relative to what, residential housing? Aluminum smelting?
It's fine to be skeptical of the AI bubble, but this article (ironic for FT) reads like veiled de-development.
We also don't need any fizzy-drinks - shall we calculate the deaths due to pollution caused by that, let alone the primary effect on health of drink-victims? If you don't like what electricity is being used for, argue for specific regulation to make it more expensive.
It’s deflection from the real polluters
Cars and suburban sprawl
> but this article (ironic for FT) reads like veiled de-development.
Hard to tell if they're singling out AI because it's the current hot thing that everyone's talking about (and afraid of being replaced by), or if they're singling out "big tech" because they're rich and cool to hate on.
> reads veiled de-development.
> If you don't like what electricity is being used for, argue for specific regulation to make it more expensive.
That reads like veiled de-development.
How do you reconcile saying both those things?
De-development is saying we have to e.g. stop driving cars. Regulations that make dirty power more expensive are the likes of a carbon tax, which you can avoid while still having a car by getting an electric car and charging it from solar panels.
Agreed. It reminds me of the article about how self-driving cars, by being safer, would worsen the organ waitlist.[1]
Like, in what other context would you raise this concern about saving lives?
It's the same kind of dishonesty, picking out one hot-button source of pollution, apportioning out its health impact, and complaining about it alone.
[1] https://slate.com/technology/2016/12/self-driving-cars-will-...
As someone who got from an organ transplant, I'd still push for self-driving car and safer cars. My hope is pig organ transplants make it through trials soon. That could push a 7 year wait for kidney transplant down to a year which is a big deal because the long time on dialysis can have a lasting impact.
In the future blinking with cost trillions of dollars.
Thanks! the captcha is annoying tho
We should be building far more nuclear.
We wildly overengineered on safety regulations making them far more costly and risky to build than they actually need to be.
Even our older Gen II nuclear reactors from the 1990's are orders of magnitude safer than current fossil fuel power stations, in terms of both released radiation and lost human lifespan years.
Moreover, the higher electricity prices that resulted from using non-nuclear power lead to ICE vehicles remaining more prevalent than they should be, and non-heat-pump heating remaining more prevalent than it should be, leading to millions more years of human lifespan lost.
From a health perspective, blocking nuclear with overregulation was one of societies biggest health policy failures.
How about we build this stuff in deserts where you can just plaster everything with solar in like 10% of time and 30% of money investment.
You now have 70% of budget to build battery/wind and still got enough time.
This is solved with current renewable prices without ideological preferences.
Unfortunately, data centers also need access to considerable amounts of water.
When a major nuclear disaster happens, it can cause hundreds of billions of damage. To make it worse from a liability perspective, that damage can be attributed to a specific power plant.
We know from natural disasters that insurance is usually insufficient to cover the true damage. Either governments have to step in, or the victims have to bear the costs. If the government has to pay for damages but someone else pays for safety features, there are clear incentives for overregulation.
The company operating the reactor obviously cannot pay. As long as shareholders are shielded from liability, it's easy enough to structure the company to ensure that it has no major assets beyond the power plant.
It's overall beneficial for society to switch to nuclear.
The problem you posed is just a market structure implementation detail. Government could operate the plants, or they could offer insurance to private sector operators.
I'm pro-nuclear but the idea that insurance fixes everything is strange to me. How much insurance money fixes you getting killed, would you say?
This is not a novel question.
If someone gets killed at work, how much do they get from workers' comp? What is a micromort and how much does one usually sell for? What did weregeld used to be back when it was still a thing?
I didn't ask about an unspecified someone else, I asked about yourself. It makes a big difference to understand what I meant.
I missed where they say insurance solves everything. I understood it as the government running it solves many of the issues introduced by the private sector, and the insurance was another thought.
Nuclear is not economically viable compared to renewables (wind & solar), and it's rightfully on its way out.
What is really needed is for humanity to say "enough" and stop using fossil fuels. Would it be a problem if training AI models (or running any other analytics) be lot more expensive in winter than summer? I doubt it. There is no reason not to use more renewables for this use case, then.
Nuclear has a far lower LCOE than renewables, when you mass produce them rather than do ad-hoc bespoke builds.
For example, the Gen II+ plants built in the 90's had an (inflation adjusted) price of about $500 million per GWe. Whereas, the cost of the recent Vogtl build was $15 billion per GWe!
Renewables proponents point to the cost of Vogtl (or similar) when they say that Nuclear has a higher cost than renewables, which is dishonest. The incremental cost of new reactors is far lower when you build many of them. The recent projects are mostly learning curve costs because of how long it's been.
The other part of this is that expensive silicon is not going to sit idle. It needs baseload power, and for solar that means adding batteries, and also downrating the nameplate capacity to it's wintertime capacity. These factors increase the real world cost of renewable power significantly.
Don't get me wrong, I think we should be investing in renewable as well, but it's still a substantially more expensive way to generate power than a serious roll out of nuclear would be.
What do you mean over engineered? Genuinely very curious, I thought the problem boiled down essentially to 'lack of will' and that there exist very safe reactor designs.
That's certainly a funny phrase, since it ought to mean that too much engineering effort was put in to getting something right but what it's actually used to mean is that someone got the trade-offs wrong.
In this case it's being used to mean that the regulations are overly restrictive or complicated, and impose higher costs than are reasonable for the benefits they provide. (Or potentially that being overcomplicated actually makes them less effective than if they were done properly.)
Who decides they are over engineered on safety?
Didn’t we already learn better safe than sorry?
Even if you account for chernobyl and fukushima the death count from nuclear is orders of magnitudes less than the death count from coal, oil, solar, etc. We overregulated nuclear because the risk of nuclear disasters feels scarier than the real risk of deaths downstream of airborne particulates. There are tradeoffs to every safety regulation on nuclear and those tradeoffs include more climate induced disasters, a sicker population, long term higher energy costs, etc.
>nuclear disasters feels scarier
I am in favor of more nuclear power and I know very little about the field, but this argument never resonated with me.
If the Chernobyl series paints a reasonably realistic image of what could have happened had they not contained the melt down, we are talking "better part of a continent being uninhabitable for thousands of years". When that is the potential risk, comparing death counts from non-catastrophic scenarios isn't what we should be looking at.
> When that is the potential risk
It's not a realistic risk.
Chernobyl was absolute worse case. A massive reactor, with no containment structure (it was essentially in a shed), with a flammable solid state moderator, and no passive cooling. Yet, Europe is not uninhabitable today, infact the above-natural-background radiation in Europe is almost entirely coming from burning fossil fuels, not from Chernobyl.
Tell us more about the death count from solar. Is it all people falling off roofs?
Pretty much yea. So not exactly a fair comparison because e.g. grid level solar doesn't necessarily involve roof climbing.
But given current installation base, it's like 3x more deaths per unit energy than wind.
I don't know how reliable it is, but the first link I clicked on contradicts your assertion: https://ourworldindata.org/grapher/death-rates-from-energy-p...
Specifically, the current regulatory framework requires not just minimization of accident, but also minimization of radiation (to levels well below that of other powerplant types like coal). There probably could be a pretty notable cost savings in nuclear plant construction if they were allowed to regularly leak as much radiation into the environment to that which we already consider acceptable from other power sources.
So if we just do something politically unacceptable and probably morally unacceptable then nuclear will be viable? And of course you could apply the same thing to other sources of energy and they will equally get cheaper. Wind turbines in beauty spots. Remove the air quality equipment from fossil fuel plants. Dam evey river.
"better safe than sorry" is the definition of a slippery slope.
but I'm curious: what event do you think taught us that?
IMO, there's far mode damage being done to the world by, say, mercury in batteries than by nuclear accidents (of which chernobyl is the only serious one).
$PROBLEM_2 being worse than $PROBLEM_1 does not mean $PROBLEM_1 is over regulated.
Nuclear Waste handling and disposal is a significant regulatory issue and includes many REACTIVE measures that have been hard-learned over the past century at sites like Pennsylvania, Fukushima, and Chernobyl.
This blind lashing out at any and all forms of regulation is ignorant. Each of these situations needs to be considered thoughtfully and contextually to find a balance between short term and long term objectives.
We do not thoughtfully balance regulatory costs on nuclear vs say coal... Nuclear regulatory compliance is orders of magnitude more expensive despite being orders of magnitude safer for human health (on top of the climate).
90% of new electricity generation in the US is solar or wind.
Coal generation has generally trended downward since the mid 2000s.
We don't need nuclear to replace coal.
We certainly do need nuclear to replace coal.
Coal and nuclear are both reliable sources of energy, unlike solar or wind. I'm sure you're about to go into some battery related hand waving, but that doesn't help in two ways. First, there will be intervals where the lull in production exceeds storage capacity, which means fatalities in many cases. Second, battery is expensive, so the total system cost can exceed the cost of nuclear (especially with sane regulation and continually improving engineering).
Also, as far as "coal trending downward", that isn't the case outside the US.
We're going to need a vastly increased energy supply in order to meet the world's needs. Nuclear fission absolutely needs to be part of that mix, as we work to master fusion and other advanced, safe, and environmentally responsible forms of energy production.
Beyond that, high density energy sources are highly desirable off-planet. :-)
Since every activity has some externalities and also unexpected positive effects, isn’t focusing on just the negative effects of this one thing very lopsided?
Now imagine how much pollution from the airline industry costs public health ...
There’s nothing specific to data centers here, just electricity use. The electricity use generates a shit load of GDP and soft power for the U.S., way more than $5.4B over 5 years.
So what do these degrowth people want exactly? Cant have resource extraction. Can’t have manufacturing, pollution. Can’t have tech, pollution. Just twiddling thumbs and money will come? Or keep all the benefits while offloading all the downsides to Mexico? And a few years later they will complain about dependence on foreign countries.
They are the people who think only one move in chess. I hope this answers your question.
What about obesity?
And now you know why big tech was so quick to kiss Trumps ring, and dump DEI. Their alliance with the left was no longer useful to them.
Elon Musk once again showed his forethought and boldness in seizing first mover advantage and going all in on this.
It’s about money and power (power to make more money) for big tech. Such was the case when they aligned with the left and it is when they court the right. Don’t take it personally and don’t shoot the messenger.
What does this have to do with the article?
Big tech is building data centres. Data centres pollute. Environmentalists are part of the left wing coalition.