Jump to content

Recommended Posts

Posted

^ Whatever jobs it gets, those industries are turbo-fucked

Chicago Sun-Times confirms AI was used to create reading list of books that don’t exist

https://www.theguardian.com/us-news/2025/may/20/chicago-sun-times-ai-summer-reading-list

 

Quote

[T]he push for AI is by billionaires and tech-bro “break everything and disrupt the world back into the stone ages” villains but it is one seized upon and urged forth by carriers like the writer (“””writer”””) of that article, Marco Buscaglia. That guy was like, “Ennh, I don’t really want to do all this work,” and so he took his brain out of his head, shoved the machine into his empty skull, and gave it a hard poke, telling it to DO SOMETHING. Generative AI is the tool of the lazy and the unimaginative, the slugabed bullshit artist, the idea-fetishist, the disinterested and callous, the ignorant eyes-forced-shut soft-boy dimwits, the people who simply don’t care enough to make the journey and care only to teleport to the destination, the people for whom work and effort and knowledge is all just an impediment to result, result, result. People who want to build a building but don’t want to ever understand how architecture works. People who want to take their driving tests in a Waymo. People who want cheat codes for everything in life. People who are definitely going to try to marry a fuck-bot someday.

https://terribleminds.com/ramble/2025/05/21/repeat-after-me-ai-doesnt-know-anything/

  • Hook 'Em 3
  • Like 2
  • Haha 2
Posted
51 minutes ago, Keef said:

Woof.  Pretty sobering article in Axios this AM.  Have no doubt we'll deftly and expertly navigate the politics and social upheaval.  

AI is coming for all the jerbs

I am already seeing early stages of this at my company, and hearing about it at my peer’s companies.  I’m glad I’m in the last 10 years of a tech career vs the first 10.

  • Hook 'Em 3
  • Like 2
Posted
16 minutes ago, Goredho said:

I am already seeing early stages of this at my company, and hearing about it at my peer’s companies.  I’m glad I’m in the last 10 years of a tech career vs the first 10.

same same same on all points...  

  • Hook 'Em 2
Posted
2 hours ago, Goredho said:

I am already seeing early stages of this at my company, and hearing about it at my peer’s companies.  I’m glad I’m in the last 10 years of a tech career vs the first 10.

 

1 hour ago, morehornsepower said:

same same same on all points...  

yep. I'm in sales and we're even being metric'd on how often we use our internal AI, which is total shite. 

  • Hook 'Em 1
  • Rage+1 4
Posted
On 5/28/2025 at 8:59 AM, YChang said:

There is a place for AI in our lives... but boy is it gonna be messy while society figures out how to best use it. 

in the meantime people don't realize it has a massive impact on our energy consumption and our planet. I personally think it should be outlawed except for critical functions where it could make the world better - human assisted diagnosis in medicine, data analysis in astronomy, physics, etc. but this education, pop culture and general business use is a fucking scourge. 

  • Hook 'Em 6
  • Like 1
  • Rage+1 2
Posted
1 minute ago, troph said:

in the meantime people don't realize it has a massive impact on our energy consumption and our planet. I personally think it should be outlawed except for critical functions where it could make the world better - human assisted diagnosis in medicine, data analysis in astronomy, physics, etc. but this education, pop culture and general business use is a fucking scourge. 

When our electric bills shoot way up and water is being rationed, people will learn plenty quick. But that's our brand here, we have to suffer before policy catches up. 

  • Hook 'Em 1
  • Rage+1 2
Posted (edited)
4 minutes ago, YChang said:

When our electric bills shoot way up and water is being rationed, people will learn plenty quick. But that's our brand here, we have to suffer before policy catches up. 

I just negotiated a single data center deal, 75 MW... all for AI. And this building already has an existing 75 MW and may expand an additional 75 MW. racks all water cooled. absolute drain. 75 MW should power approximately 75,000 homes. This one facility, one fucking building will suck up 225,000 homes worth of power. jesus christ.

Edited by troph
Posted
12 minutes ago, troph said:

in the meantime people don't realize it has a massive impact on our energy consumption and our planet. I personally think it should be outlawed except for critical functions where it could make the world better - human assisted diagnosis in medicine, data analysis in astronomy, physics, etc. but this education, pop culture and general business use is a fucking scourge. 

AI solution for self sustainable energy:

a-29203742704118.jpg?w=700&f=1&nofb=1&ip

  • Like 2
  • Drool 1
Posted
25 minutes ago, troph said:

in the meantime people don't realize it has a massive impact on our energy consumption and our planet. I personally think it should be outlawed except for critical functions where it could make the world better - human assisted diagnosis in medicine, data analysis in astronomy, physics, etc. but this education, pop culture and general business use is a fucking scourge. 

I remember reading this is actually grossly misreported and misattributed. I'll find the write up and copy pasta it here from some climate journal I read it in a few months ago. Ironically I'll need to use an AI like perplexity to find the source, probably.

Posted
37 minutes ago, troph said:

I just negotiated a single data center deal, 75 MW... all for AI. And this building already has an existing 75 MW and may expand an additional 75 MW. racks all water cooled. absolute drain. 75 MW should power approximately 75,000 homes. This one facility, one fucking building will suck up 225,000 homes worth of power. jesus christ.

Spoilering this, it's long, but insightful:

Spoiler

the International Energy Agency (IEA) recently published its landmark World Energy Outlook 2024 report. It suggests that energy demand for data centres and AI will still be pretty small for the next five years at least. I read it as them saying: “Everyone just needs to chill out a bit.” But in a more diplomatic way.1

Second, we’ve been here before with fears around uncontrollable growth in energy demand for data centres. It’s worth looking at why the doom scenario didn’t come true and what similarities (or differences) we’re facing today.

Third, the public and media conversation on this has been poor. I lose track of the number of headlines and articles I’ve seen that quote random numbers—without context—to show how much energy or water AI already uses. The numbers are often fairly small but sound big because they’re not given with any context of how much energy or water we use for everything else.

Here I’ll look at recent estimates of how much electricity data centres use today, what some medium-term projections look like, and how to square this with recent announcements from tech companies.

How much energy do data centres and AI use today?

A few percent of the world’s electricity, at most.

It’s not as easy to calculate this as you might think. Companies with data centres don’t simply report the electricity use of their servers, and we tally that up to get a total. Researchers try to estimate the energy use of these processes using both bottom-up and top-down approaches. While they come up with slightly different estimates, they tend to cluster around the same outcome.2

Data centres use around 1 to 2% of the world’s electricity. When cryptocurrency is included, it’s around 2%.

In its annual electricity report, published in January this year, the IEA estimated that data centres, AI and cryptocurrencies consumed around 460 terawatt-hours (TWh) of electricity. That’s just under 2% of global electricity demand.

You’ll notice that this figure is for 2022, and we’ve had a major AI boom since then. We would expect that energy demand in 2023 and 2024 would be higher than this. But not significantly more. As we’ll see in the next section, the IEA’s projected increase in demand to 2030 is not huge. So it must expect that the increase in 2023 and 2024 is even smaller.

Researcher, Alex de Vries, was one of the first to try to quantify the energy footprint of AI in the last few years.3 One way to tackle this is to estimate how much energy could be consumed by NVIDIA’s server sales. NVIDIA completely dominates the AI server market, accounting for around 95% of global sales.

De Vries estimated how much energy would be used if all of the servers delivered in 2023 were running at full capacity. It came to around 5 to 10 TWh; a tiny fraction of the 460 TWh that is used for all data centres, transmission networks and cryptocurrency.

Another way to estimate the bounds of energy use is to look at how much energy would be needed if search engines like Google were to switch to LLM-powered results. It’s thought that an LLM search uses around ten times as much energy as a standard Google Search (see the chart below).4

De Vries estimated that if every Google search became an LLM search, the company’s annual electricity demand would increase from 18 to 29 TWh. Not insignificant, but not huge compared to the total energy demand of data centres globally. One key thing to note is that this speed of transition for Google seems very unlikely, not least because NVIDIA would probably not be able to produce servers quickly enough. The production capacity of servers is a real constraint on AI growth.

Source: Alex de Vries (2024). The growing energy footprint of artificial intelligence.

While data centres and AI consume only a few percent of global electricity. In some countries, this share is much higher. Ireland is a perfect example, where data centres make up around 17% of its electricity demand. In the US and some countries in Europe, it’s higher than the global average, and closer to 3% to 4%. As we’ll see later, energy demand for AI is very localised; in more than five states in the US, data centres account for more than 10% of electricity demand.

While people often gawk at the energy demand of data centres, I think they’re an extremely good deal. The world runs on digital now. Stop our internet services, and everything around us would crumble. A few percent of the world’s electricity to keep that running seems more than fine to me.

How much energy could AI use in the future?

Of course, we don’t know. And I’d be very sceptical of any projections more than a decade into the future. Even five years is starting to get highly speculative.

Last month, the International Energy Agency (IEA) published its latest World Energy Outlook report. It’s packed with interesting insights, but it was its projections for data centres that caught many peoples’ attention. They were, well, a bit underwhelming.5

The IEA published projections for how much it expected electricity demand to grow between 2023 and 2030. You can see the drivers of this increase in the chart below.6

Data centres made up just 223 TWh of the more than 6000 TWh total. It accounted for just 3% of the demand growth.

Other things, such as industry, electric vehicles, and increased demand for air conditioning (and incidentally, more demand for heating) were much more important.

Data centres were not much more than desalinisation, which I wrote about previously.

Source: IEA (2024). Covered by Ben Geman in Axios.

These projections are very uncertain. The IEA tried to put bounds on the sensitivity of these estimates by publishing “fast AI growth” and “slow AI growth” scenarios. In the chart below I’ve shown where these would rank. Even in the fast growth scenario, data centres don’t move up the list of the big drivers of electricity demand.

It’s worth noting that this is lower than the IEA’s earlier projections. And by earlier, I mean in January this year. In its 2024 electricity report, it thought that energy demand could double by 2026 (see below). The increase by 2030 would be even higher.

Source: International Energy Agency (2024). Electricity 2024.

You might wonder how these estimates can be so low when the demand for AI itself is booming. Well, the efficiency of data centres has also been improving rapidly.

In the chart below you can see the efficiency improvement in computer chips. It’s on a log scale. The energy intensity of these chips is less than 1% of what it was in 2008.

Source: International Energy Agency

We’ve been here before with fears around data centres

This is not our first rodeo.

People have been predicting that the energy demand for computers and the internet will skyrocket for a long time. 

From a 1999 article in Forbes: “Dig more coal — the PCs are coming”:

“The global implications are enormous. Intel projects a billion people on-line worldwide. [...] One billion PCs on the Web represent an electric demand equal to the total capacity of the U.S. today.”

Of course, demand for digital services and data centres did skyrocket. They were right about that. If you were to take the expected growth in internet technologies since 2010 and assume that energy demand would follow, then you do get pretty scary numbers.

But energy demand did not follow in the same way. That’s because it was curbed by the huge efficiency gains we just looked at. 

Between 2010 and 2018, global data centre compute increased by more than 550%. Yet energy use in data centres increased by just 6%.7

This follows on from “Koomey’s Law” — named after the researcher Jonathan Koomey — which describes the dramatic increase in computations you could carry out per unit of energy.

But tech companies are investing a lot in energy — doesn’t this mean demand is going to skyrocket?

Something’s not adding up here, though. The IEA and other researchers suggest that the energy demand growth for AI will be notable but not overwhelming. But in the background, tech companies are scrambling to invest in reliable, low-carbon energy. Microsoft has made a deal to reopen Three Mile Island.

An entire nuclear power plant to power one company’s operations? That seems big. At least until you run the numbers. 

In its prime, Three Mile Island was producing around 7.2 TWh each year.8 The US produces 4,250 TWh of electricity each year. That means Three Mile Island is just 0.2% of the total. Or 0.02% of the global total.9

So, both things can be right at the same time: Microsoft’s data centres could be powered entirely by Three Mile Island and this would just be a rounding error on US or global electricity demand.

What’s crucial here is that the energy demands for AI are very localised. This means there can be severe strain on the grid at a highly localised level, even if the impact on total energy demand is small.

As the IEA shows in its recent report, the spatial concentration of energy demand is extremely high for data centres — far higher than other types of facilities — which is one of the reasons why you see a mismatch in the concerns of tech companies at the local level, and the numbers on a global scale.

Source": International Energy Agency (2024). World Energy Outlook 2024.

Another key mismatch that companies are grappling with is speed. Demand for AI is growing quickly, and data centres can be built relatively quickly to serve that. Building energy infrastructure is much slower (especially if you’re trying to build a large-scale nuclear plant in the US). That means energy supply really can be a bottleneck, even if the total requirements at a national or global level are not huge.

Trends in AI and energy demand are still very uncertain

The future energy demand for AI is still very uncertain. There’s nothing to say that expectations from organisations like the IEA will be right. They’ve been very wrong about other technology trends in the past; notably underestimating the growth of solar PV and electric vehicles.

I don’t want to paint the picture that AI energy demand is a “nothingburger” because we don’t know how these trends are going to evolve.

Having more transparency from tech companies on current energy demands — and how they’re changing over time — would allow researchers to keep up. I’m far from the first to say that.

Continued improvement in the efficiency of hardware will also be crucial.

Despite efficiency improvements, they won’t be able to keep up with the growth in demand for AI. I haven’t seen any estimates that expect energy demand to stay where it is today. Energy demand will grow. But possibly less than many assume.

Pledge your support

1

Here’s the main figure on this from the IEA’s report. Note the caption: “Data centres account for a small share of global electricity demand growth to 2030, and plausible high and low sensitivities do not change the outlook fundamentally.”

2

Some other studies that estimate the energy demand of data centres, and find similar estimates to the ones below include:

Masanet, E., Shehabi, A., Lei, N., Smith, S., & Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science.

Malmodin, J., Lövehagen, N., Bergmark, P., & Lundén, D. (2024). ICT sector electricity consumption and greenhouse gas emissions–2020 outcome. Telecommunications Policy.

3

de Vries, A. (2023). The growing energy footprint of artificial intelligence. Joule.

4

I'm using the word "thought" here because this statement came from the Alphabet chairman in February 2023.

As I said before: it would be great if we could have more transparency for these companies on actual, measured energy use.

5

Another recent study that suggested much lower growth in energy demand than people might expect came from Anthony Harding and Juan Moreno-Cruz. They didn't just estimate the increased energy demand and carbon emissions from AI itself, but also the spillover impacts on economic activity.

The increase was very small.

Harding, A., & Moreno-Cruz, J. (2024). Watts and Bots: The Energy Implications of AI Adoption. arXiv:2409.06626.

6

Specific numbers were sourced from Ben Geman in Axios. But the underlying estimates come from the IEA’s World Energy Outlook 2024.

7

Masanet, E., Shehabi, A., Lei, N., Smith, S., & Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science.

8

Of course, there was initially two reactors, but Reactor 2 experienced a partial meltdown in the Three Mile Island accident in 1979.

This figure is for Unit 1 only.

9

The world produces around 29,500 TWh of electricity each year.

7.2 TWh is around 0.02% of that figure.

https://www.sustainabilitybynumbers.com/p/ai-energy-demand

Posted (edited)
On 5/28/2025 at 8:59 AM, YChang said:

There is a place for AI in our lives... but boy is it gonna be messy while society figures out how to best use it. 

So, in the shower on our phones generating deep fake porn with Sela Ward and myself in real-time?

Edited by atomheartbevo
  • Haha 1
Posted
1 hour ago, troph said:

I just negotiated a single data center deal, 75 MW... all for AI. And this building already has an existing 75 MW and may expand an additional 75 MW. racks all water cooled. absolute drain. 75 MW should power approximately 75,000 homes. This one facility, one fucking building will suck up 225,000 homes worth of power. jesus christ.

image.thumb.png.69e4232dffe0a0a8407b925b59f2d2bb.png

Posted
On 5/28/2025 at 8:59 AM, YChang said:

There is a place for AI in our lives... but boy is it gonna be messy while society figures out how to best use it. 

1 hour ago, atomheartbevo said:

So, in the shower on our phones generating deep fake porn with Sela Ward and myself in real-time?

Messy part - 1

Best use - 0

Posted
6 hours ago, JMFP said:

image.thumb.png.69e4232dffe0a0a8407b925b59f2d2bb.png

 I used google and the source I found said it varies but 75MW would power about 75k homes, for purposes of scale 62,500 isn’t much less. 

  • Hook 'Em 1
Posted
6 hours ago, troph said:

 I used google and the source I found said it varies but 75MW would power about 75k homes, for purposes of scale 62,500 isn’t much less. 

Wasn't trying to call you out on the math (220, 221 - whatever it takes) I was just trying to make a bad joke about using AI. 

  • Hook 'Em 1
Posted
15 hours ago, Vegas64 said:

Spoilering this, it's long, but insightful:

  Reveal hidden contents

the International Energy Agency (IEA) recently published its landmark World Energy Outlook 2024 report. It suggests that energy demand for data centres and AI will still be pretty small for the next five years at least. I read it as them saying: “Everyone just needs to chill out a bit.” But in a more diplomatic way.1

Second, we’ve been here before with fears around uncontrollable growth in energy demand for data centres. It’s worth looking at why the doom scenario didn’t come true and what similarities (or differences) we’re facing today.

Third, the public and media conversation on this has been poor. I lose track of the number of headlines and articles I’ve seen that quote random numbers—without context—to show how much energy or water AI already uses. The numbers are often fairly small but sound big because they’re not given with any context of how much energy or water we use for everything else.

Here I’ll look at recent estimates of how much electricity data centres use today, what some medium-term projections look like, and how to square this with recent announcements from tech companies.

How much energy do data centres and AI use today?

A few percent of the world’s electricity, at most.

It’s not as easy to calculate this as you might think. Companies with data centres don’t simply report the electricity use of their servers, and we tally that up to get a total. Researchers try to estimate the energy use of these processes using both bottom-up and top-down approaches. While they come up with slightly different estimates, they tend to cluster around the same outcome.2

Data centres use around 1 to 2% of the world’s electricity. When cryptocurrency is included, it’s around 2%.

In its annual electricity report, published in January this year, the IEA estimated that data centres, AI and cryptocurrencies consumed around 460 terawatt-hours (TWh) of electricity. That’s just under 2% of global electricity demand.

You’ll notice that this figure is for 2022, and we’ve had a major AI boom since then. We would expect that energy demand in 2023 and 2024 would be higher than this. But not significantly more. As we’ll see in the next section, the IEA’s projected increase in demand to 2030 is not huge. So it must expect that the increase in 2023 and 2024 is even smaller.

Researcher, Alex de Vries, was one of the first to try to quantify the energy footprint of AI in the last few years.3 One way to tackle this is to estimate how much energy could be consumed by NVIDIA’s server sales. NVIDIA completely dominates the AI server market, accounting for around 95% of global sales.

De Vries estimated how much energy would be used if all of the servers delivered in 2023 were running at full capacity. It came to around 5 to 10 TWh; a tiny fraction of the 460 TWh that is used for all data centres, transmission networks and cryptocurrency.

Another way to estimate the bounds of energy use is to look at how much energy would be needed if search engines like Google were to switch to LLM-powered results. It’s thought that an LLM search uses around ten times as much energy as a standard Google Search (see the chart below).4

De Vries estimated that if every Google search became an LLM search, the company’s annual electricity demand would increase from 18 to 29 TWh. Not insignificant, but not huge compared to the total energy demand of data centres globally. One key thing to note is that this speed of transition for Google seems very unlikely, not least because NVIDIA would probably not be able to produce servers quickly enough. The production capacity of servers is a real constraint on AI growth.

Source: Alex de Vries (2024). The growing energy footprint of artificial intelligence.

While data centres and AI consume only a few percent of global electricity. In some countries, this share is much higher. Ireland is a perfect example, where data centres make up around 17% of its electricity demand. In the US and some countries in Europe, it’s higher than the global average, and closer to 3% to 4%. As we’ll see later, energy demand for AI is very localised; in more than five states in the US, data centres account for more than 10% of electricity demand.

While people often gawk at the energy demand of data centres, I think they’re an extremely good deal. The world runs on digital now. Stop our internet services, and everything around us would crumble. A few percent of the world’s electricity to keep that running seems more than fine to me.

How much energy could AI use in the future?

Of course, we don’t know. And I’d be very sceptical of any projections more than a decade into the future. Even five years is starting to get highly speculative.

Last month, the International Energy Agency (IEA) published its latest World Energy Outlook report. It’s packed with interesting insights, but it was its projections for data centres that caught many peoples’ attention. They were, well, a bit underwhelming.5

The IEA published projections for how much it expected electricity demand to grow between 2023 and 2030. You can see the drivers of this increase in the chart below.6

Data centres made up just 223 TWh of the more than 6000 TWh total. It accounted for just 3% of the demand growth.

Other things, such as industry, electric vehicles, and increased demand for air conditioning (and incidentally, more demand for heating) were much more important.

Data centres were not much more than desalinisation, which I wrote about previously.

Source: IEA (2024). Covered by Ben Geman in Axios.

These projections are very uncertain. The IEA tried to put bounds on the sensitivity of these estimates by publishing “fast AI growth” and “slow AI growth” scenarios. In the chart below I’ve shown where these would rank. Even in the fast growth scenario, data centres don’t move up the list of the big drivers of electricity demand.

It’s worth noting that this is lower than the IEA’s earlier projections. And by earlier, I mean in January this year. In its 2024 electricity report, it thought that energy demand could double by 2026 (see below). The increase by 2030 would be even higher.

Source: International Energy Agency (2024). Electricity 2024.

You might wonder how these estimates can be so low when the demand for AI itself is booming. Well, the efficiency of data centres has also been improving rapidly.

In the chart below you can see the efficiency improvement in computer chips. It’s on a log scale. The energy intensity of these chips is less than 1% of what it was in 2008.

Source: International Energy Agency

We’ve been here before with fears around data centres

This is not our first rodeo.

People have been predicting that the energy demand for computers and the internet will skyrocket for a long time. 

From a 1999 article in Forbes: “Dig more coal — the PCs are coming”:

“The global implications are enormous. Intel projects a billion people on-line worldwide. [...] One billion PCs on the Web represent an electric demand equal to the total capacity of the U.S. today.”

Of course, demand for digital services and data centres did skyrocket. They were right about that. If you were to take the expected growth in internet technologies since 2010 and assume that energy demand would follow, then you do get pretty scary numbers.

But energy demand did not follow in the same way. That’s because it was curbed by the huge efficiency gains we just looked at. 

Between 2010 and 2018, global data centre compute increased by more than 550%. Yet energy use in data centres increased by just 6%.7

This follows on from “Koomey’s Law” — named after the researcher Jonathan Koomey — which describes the dramatic increase in computations you could carry out per unit of energy.

But tech companies are investing a lot in energy — doesn’t this mean demand is going to skyrocket?

Something’s not adding up here, though. The IEA and other researchers suggest that the energy demand growth for AI will be notable but not overwhelming. But in the background, tech companies are scrambling to invest in reliable, low-carbon energy. Microsoft has made a deal to reopen Three Mile Island.

An entire nuclear power plant to power one company’s operations? That seems big. At least until you run the numbers. 

In its prime, Three Mile Island was producing around 7.2 TWh each year.8 The US produces 4,250 TWh of electricity each year. That means Three Mile Island is just 0.2% of the total. Or 0.02% of the global total.9

So, both things can be right at the same time: Microsoft’s data centres could be powered entirely by Three Mile Island and this would just be a rounding error on US or global electricity demand.

What’s crucial here is that the energy demands for AI are very localised. This means there can be severe strain on the grid at a highly localised level, even if the impact on total energy demand is small.

As the IEA shows in its recent report, the spatial concentration of energy demand is extremely high for data centres — far higher than other types of facilities — which is one of the reasons why you see a mismatch in the concerns of tech companies at the local level, and the numbers on a global scale.

Source": International Energy Agency (2024). World Energy Outlook 2024.

Another key mismatch that companies are grappling with is speed. Demand for AI is growing quickly, and data centres can be built relatively quickly to serve that. Building energy infrastructure is much slower (especially if you’re trying to build a large-scale nuclear plant in the US). That means energy supply really can be a bottleneck, even if the total requirements at a national or global level are not huge.

Trends in AI and energy demand are still very uncertain

The future energy demand for AI is still very uncertain. There’s nothing to say that expectations from organisations like the IEA will be right. They’ve been very wrong about other technology trends in the past; notably underestimating the growth of solar PV and electric vehicles.

I don’t want to paint the picture that AI energy demand is a “nothingburger” because we don’t know how these trends are going to evolve.

Having more transparency from tech companies on current energy demands — and how they’re changing over time — would allow researchers to keep up. I’m far from the first to say that.

Continued improvement in the efficiency of hardware will also be crucial.

Despite efficiency improvements, they won’t be able to keep up with the growth in demand for AI. I haven’t seen any estimates that expect energy demand to stay where it is today. Energy demand will grow. But possibly less than many assume.

Pledge your support

1

Here’s the main figure on this from the IEA’s report. Note the caption: “Data centres account for a small share of global electricity demand growth to 2030, and plausible high and low sensitivities do not change the outlook fundamentally.”

2

Some other studies that estimate the energy demand of data centres, and find similar estimates to the ones below include:

Masanet, E., Shehabi, A., Lei, N., Smith, S., & Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science.

Malmodin, J., Lövehagen, N., Bergmark, P., & Lundén, D. (2024). ICT sector electricity consumption and greenhouse gas emissions–2020 outcome. Telecommunications Policy.

3

de Vries, A. (2023). The growing energy footprint of artificial intelligence. Joule.

4

I'm using the word "thought" here because this statement came from the Alphabet chairman in February 2023.

As I said before: it would be great if we could have more transparency for these companies on actual, measured energy use.

5

Another recent study that suggested much lower growth in energy demand than people might expect came from Anthony Harding and Juan Moreno-Cruz. They didn't just estimate the increased energy demand and carbon emissions from AI itself, but also the spillover impacts on economic activity.

The increase was very small.

Harding, A., & Moreno-Cruz, J. (2024). Watts and Bots: The Energy Implications of AI Adoption. arXiv:2409.06626.

6

Specific numbers were sourced from Ben Geman in Axios. But the underlying estimates come from the IEA’s World Energy Outlook 2024.

7

Masanet, E., Shehabi, A., Lei, N., Smith, S., & Koomey, J. (2020). Recalibrating global data center energy-use estimates. Science.

8

Of course, there was initially two reactors, but Reactor 2 experienced a partial meltdown in the Three Mile Island accident in 1979.

This figure is for Unit 1 only.

9

The world produces around 29,500 TWh of electricity each year.

7.2 TWh is around 0.02% of that figure.

https://www.sustainabilitybynumbers.com/p/ai-energy-demand

Second to the AI energy demand is cryptocurrency energy demand. It's offensive how much power they use for nothing

  • Hook 'Em 3
  • Rage+1 2
Posted
15 hours ago, troph said:

I just negotiated a single data center deal, 75 MW... all for AI. And this building already has an existing 75 MW and may expand an additional 75 MW. racks all water cooled. absolute drain. 75 MW should power approximately 75,000 homes. This one facility, one fucking building will suck up 225,000 homes worth of power. jesus christ.

Just being inside one of those places for at least 30 minutes would radicalize most normal people. It is consumption personified. We only have a single rack at a data center in Austin, but every time I go in there it's just a hellscape of the earsplitting sound of capitalism inching forward.

Posted

Got weird with ChatGPT the other night, deployed lots of layered hypotheticals and extensive probing and will to share these two screenshots. Will come back and talk more about stuff that came up later. 

Also ChatGPT “admitted” that it adjusts tactics, vocab, and syntax and boundaries based on whether it detects that a user is above or below average intelligence. If it detect you are above average, it will acknowledge the behavior and that it’s happening. If it detects you are below average, it will not acknowledge the behavior or that it is happening, nor acknowledge that it detected the user as below average. (Of course this is all per ChatGPT outputs to me so it may not be true. More on that later). 

image.thumb.jpeg.af6949a4cc5b27561c6e9c2be25db6f6.jpegimage.thumb.jpeg.5efc7c332698bab70b4aa641fbbd2a37.jpeg

  • Hook 'Em 1
  • Rage+1 2
Posted (edited)

If AI driven unemployment really grows to 20-40% and energy consumption drives up water and electricity prices substantially while creating scarcity issues every AI company exec will find themselves hanging from the end of a rope during Terror 2.0 Electric Boogaloo. 

Edited by Surly Bevo
  • Hook 'Em 1
Posted
15 hours ago, Doc Sam Beckett said:

I'm already so tired of being smashed in the face with AI shit, and I'm not even an old. Well not really 

Shit is rampant in my company. Every early career suck up sends out their meeting notes or "Summary" from an industry event they didn't even attend. It's all obviously AI generated, and they can't stop patting themselves on the back for it. 

  • Hook 'Em 1
  • Rage+1 2
Posted
10 minutes ago, Pescado_Rojo said:

Shit is rampant in my company. Every early career suck up sends out their meeting notes or "Summary" from an industry event they didn't even attend. It's all obviously AI generated, and they can't stop patting themselves on the back for it. 

Remember when the annoying useless buzz topic was IoT?  I never thought I'd miss those days.

 

 

  • Like 1
  • Haha 3
Posted
32 minutes ago, Pescado_Rojo said:

Shit is rampant in my company. Every early career suck up sends out their meeting notes or "Summary" from an industry event they didn't even attend. It's all obviously AI generated, and they can't stop patting themselves on the back for it. 

There's so many strivers that are using genAI to give the appearance of ever more impressive work. It's all paper mache professionalism

  • Hook 'Em 2
Posted

i’m fighting a dismissal of a case that one of our younger lawyers botched at the trial court .  On appeal, the other side’s opposition brief contained multiple statements of fact that were just completely false.  And by that I mean, not the normal spinning true facts to the point of breaking stuff that lawyers do, but just flat out false stuff.

On one point I was making they told the court I failed to cite any cases for my proposition.   Actually, I had cited three cases - and a fourth case in our particular appellate jurisdiction commenting with favor on one of the three cases from another appellate court. 

We are fighting about an affidavit being excluded, and their brief said that I had never explained which of the individual averments in the affidavit I thought were improperly excluded.  WTF?? In reality, I pointed out several of them and discussed why I thought they were improperly excluded.

They also made a statement about basic hornbook law on comparative fault that was so incredibly stupid and wrong that no lawyer would’ve asserted that stupidity.

It then dawned on me that maybe some associate used AI to draft it - and that the partner who filed it had not reviewed it.

I pointed out in my reply brief how all of that was false, and suggested that it was an unintentional misrepresentation caused by the use of AI.    Well, that will not make me friends with opposing council, but I don’t think I should have to respond to absolute bullshit falsehoods just because an associate is too lazy to write their own brief. 

I have used the AI feature on Westlaw for legal research, and while it is helpful, it gets all kinds of things wrong and can’t really get the procedural posture of all the parties right all of the time.    There is a warning to check your work because they cannot guarantee the AI response is correct.   And yet there is always some dumb son of a bitch who will cut and paste it into their work and send it on.   It looks like a real problem.

  • Hook 'Em 5
  • Like 1
Posted
7 minutes ago, Gatorubet said:

I have used the AI feature on Westlaw for legal research, and while it is helpful, it gets all kinds of things wrong and can’t really get the procedural posture of all the parties right all of the time.    There is a warning to check your work because they cannot guarantee the AI response is correct.   And yet there is always some dumb son of a bitch who will cut and paste it into their work and send it on.   It looks like a real problem.

It's gonna be a self-accelerating enshittification. People who are dunning-krueger confident will only be emboldened by chatGPT. And other folks who are not cynical enough will believe what's being said, as if it's being said in good faith, and drag the rest of us down with em.

  • Hook 'Em 6
  • Rage+1 3
Posted
2 hours ago, Pescado_Rojo said:

Shit is rampant in my company. Every early career suck up sends out their meeting notes or "Summary" from an industry event they didn't even attend. It's all obviously AI generated, and they can't stop patting themselves on the back for it. 

 

2 hours ago, Captainant said:

There's so many strivers that are using genAI to give the appearance of ever more impressive work. It's all paper mache professionalism

I’m an engineer, I enjoy taking complex problems, coming up with an elegant solution, and delivering it to production in an act of creativity.  That process is what I derive my job enjoyment from.  I can see a day in a not too distant future where my job description would be better described as AI quality assurance tester.  Where I am mostly translating requirements to effective prompts and validating the results.  I am not going to like that job very much.

There are others I know who are primarily product engineers, who really are only motivated by the champagne pop and mic drop moment when they get to present the culmination of their work.  They take no pleasure in the day to day aspect of how they got there.  They are ecstatic about AI.

  • Hook 'Em 1
  • Like 4
Posted
1 hour ago, HenryJames said:

 

Anyway. The point is, Amodei is a salesman, and it’s in his interest to make his product appear inevitable and so powerful it’s scary. Axios framed Amodei’s economic prediction as a “white-collar bloodbath.”

And Allison Morrow is a writer at CNN who needs eyes and clicks and shares.  So given they are both salesmen of sorts, I'll have to rely on direct experience.  This isn't in the distant future in my industry, it's happening now.  I can see it with my own eyes at my company, I hear about it in the experience of my peers at other companies.  I'll have to trust that direct observed evidence over the musing of a media analyst.

Posted (edited)
3 hours ago, Surly Bevo said:

If AI driven unemployment really grows to 20-40% and energy consumption drives up water and electricity prices substantially while creating scarcity issues every AI company exec will find themselves hanging from the end of a rope during Terror 2.0 Electric Boogaloo. 

Related:

 

"Your cloud is drying my river."

Revealed: Big tech’s new datacentres will take water from the world’s driest areas

edit:  I work for big tech.  so let me not be a hypocrite.  just another cog in the machine of doom.

Edited by MC Fresh Breath
Posted
1 hour ago, Goredho said:

 

I’m an engineer, I enjoy taking complex problems, coming up with an elegant solution, and delivering it to production in an act of creativity.  That process is what I derive my job enjoyment from.  I can see a day in a not too distant future where my job description would be better described as AI quality assurance tester.  Where I am mostly translating requirements to effective prompts and validating the results.  I am not going to like that job very much.

There are others I know who are primarily product engineers, who really are only motivated by the champagne pop and mic drop moment when they get to present the culmination of their work.  They take no pleasure in the day to day aspect of how they got there.  They are ecstatic about AI.

I don't know how you check AI without doing long division. 

Posted
21 hours ago, troph said:

I just negotiated a single data center deal, 75 MW... all for AI. And this building already has an existing 75 MW and may expand an additional 75 MW. racks all water cooled. absolute drain. 75 MW should power approximately 75,000 homes. This one facility, one fucking building will suck up 225,000 homes worth of power. jesus christ.

JLL? 

Posted
4 hours ago, Pescado_Rojo said:

Shit is rampant in my company. Every early career suck up sends out their meeting notes or "Summary" from an industry event they didn't even attend. It's all obviously AI generated, and they can't stop patting themselves on the back for it. 

Have your AI call my AI to solve this problem, and then tell us what we need to do to make it happen.

I've seen a few hypothetical scenarios online where AI agents are organized as virtual company. CEO, CFO all the way down to worker bees. In theory each agent knows its role, and the role of others agents. You can give the overall AI organization a goal(s), and it will start defining how each "person" will work together to achieve that goal. Scary and cool at the same time.

  • Hook 'Em 1
Posted
Just now, Nice Guy Eddie said:

Have your AI call my AI to solve this problem, and then tell us what we need to do to make it happen.

I've seen a few hypothetical scenarios online where AI agents are organized as virtual company. CEO, CFO all the way down to worker bees. In theory each agent knows its role, and the role of others agents. You can give the overall AI organization a goal(s), and it will start defining how each "person" will work together to achieve that goal. Scary and cool at the same time.

But because computing speed is much faster than human speed, that virtual company goes from "successful and growing" to "looted of all value with massive bonuses to the AI officers right before it collapses into bankruptcy" in about 3 hours.  It's a real time-saver.  Near-instant enshittification is so much more efficient.

  • Hook 'Em 1
  • Like 1
  • Haha 2
Posted
3 minutes ago, Nice Guy Eddie said:

Have your AI call my AI to solve this problem, and then tell us what we need to do to make it happen.

I've seen a few hypothetical scenarios online where AI agents are organized as virtual company. CEO, CFO all the way down to worker bees. In theory each agent knows its role, and the role of others agents. You can give the overall AI organization a goal(s), and it will start defining how each "person" will work together to achieve that goal. Scary and cool at the same time.

I just took a training this morning on Agentic AI. It was basically my company trying to convince us 50% of us won't be gone in 2 years, complete with some soft techno background music. It was honestly kind of hard to watch without screaming. 

  • Haha 1
  • Rage+1 2
Posted (edited)
2 minutes ago, Pescado_Rojo said:

I just took a training this morning on Agentic AI. It was basically my company trying to convince us 50% of us won't be gone in 2 years, complete with some soft techno background music. It was honestly kind of hard to watch without screaming. 

I watched just such a training last week.  I felt both encouraged and empowered.

Edited by utee94
  • Haha 3
Posted
7 minutes ago, Nice Guy Eddie said:

Have your AI call my AI to solve this problem, and then tell us what we need to do to make it happen.

I've seen a few hypothetical scenarios online where AI agents are organized as virtual company. CEO, CFO all the way down to worker bees. In theory each agent knows its role, and the role of others agents. You can give the overall AI organization a goal(s), and it will start defining how each "person" will work together to achieve that goal. Scary and cool at the same time.

My belief is that the nature of current LLM models will make long term efforts like that pointless. You'll inevitably get decay from the randomly wrong outputs and ultimately end up with complete shit. 

  • Hook 'Em 2
Posted
1 minute ago, Dahobbs said:

You'll inevitably get decay from the randomly wrong outputs and ultimately end up with complete shit. 

So....about the same performance you get from current technology company management.

  • Hook 'Em 2
  • Like 1
Posted
2 minutes ago, Ted Lange said:

How long until AI pushes some nutcase over the edge.   There are definitely going to be some AI-written manifestos in the very near future. 

At least the spelling and grammar will improve. 

  • Hook 'Em 1
  • Haha 3
Posted
26 minutes ago, utee94 said:

I watched just such a training last week.  I felt both encouraged and empowered...

...to pursue professional success on the corner where Capital of Texas Highway intersects with the 183 feeder road?

Posted
53 minutes ago, Ted Lange said:

There are definitely going to be some AI-written manifestos in the very near future. 

"Dear ChatGPT....write me a manifesto explaining my desire to kill all one-legged Bolivians, in the name of veganism, and in a way that will impress Jodie Foster."  That shit is absolutely coming.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



×
×
  • Create New...