Imagine something that in 2020 didn't really exist but has so exploded onto our environment that in 2030 it will demand the water of 1.3 billion people and the electricity of 2 billion.
-
@gerrymcgovern
Can someone please explain to me how data centres will 'consume' so much water? There are plenty of energy intensive industries that have a heat dissipation need. If there's so much energy to dissipate put a fucking big heat exchanger out at sea and think about making them nursaries for fish/coral or something. It seems MAD to keep data centres where shifting the heat means single-use water. Use brine or molten salt or puréed billionaires in a closed loop heat exchanger near a heat sump where heat damage will be containable & acceptable.OR just say, "the price for that much AI is unsustainable. Sorry, but NO!"
@gregalotl Up until a couple of years ago, most data centers didn't even meter their water use because they were getting it so cheap. Evaporation is the cheapest way to cool and they will always go for the cheapest unless they are forced not to. They have total contempt for our environment and for the communities in which they locate.
-
@gerrymcgovern
Can someone please explain to me how data centres will 'consume' so much water? There are plenty of energy intensive industries that have a heat dissipation need. If there's so much energy to dissipate put a fucking big heat exchanger out at sea and think about making them nursaries for fish/coral or something. It seems MAD to keep data centres where shifting the heat means single-use water. Use brine or molten salt or puréed billionaires in a closed loop heat exchanger near a heat sump where heat damage will be containable & acceptable.OR just say, "the price for that much AI is unsustainable. Sorry, but NO!"
@gregalotl @gerrymcgovern because capitalists don't care that the factors of their production are also ours. "Does water cool server racks?" X 10000
-
@gerrymcgovern
Can someone please explain to me how data centres will 'consume' so much water? There are plenty of energy intensive industries that have a heat dissipation need. If there's so much energy to dissipate put a fucking big heat exchanger out at sea and think about making them nursaries for fish/coral or something. It seems MAD to keep data centres where shifting the heat means single-use water. Use brine or molten salt or puréed billionaires in a closed loop heat exchanger near a heat sump where heat damage will be containable & acceptable.OR just say, "the price for that much AI is unsustainable. Sorry, but NO!"
@gregalotl @gerrymcgovern AI is anti-human.
-
@gregalotl Up until a couple of years ago, most data centers didn't even meter their water use because they were getting it so cheap. Evaporation is the cheapest way to cool and they will always go for the cheapest unless they are forced not to. They have total contempt for our environment and for the communities in which they locate.
@gerrymcgovern
OR just say...NO!EU style legislation - freedom, but WITH RESPONSIBILITY
US, run by deluded, selfish zealots
-
Imagine something that in 2020 didn't really exist but has so exploded onto our environment that in 2030 it will demand the water of 1.3 billion people and the electricity of 2 billion.
In a time when we have a global historic severe drought and when our civilization is about to collapse due to over-consumption of energy.
This is AI.
This is Big Tech
These are the tech bros. Here to burn it all down for greed and ambition.
"AI is becoming something much larger. It is becoming one of the most important energy stories of the decade."
So says Big Oil
AI is an energy industry play.
"The scale of demand is extraordinary."
AI is a life destroyer. If you use AI, you are an active participant in the destruction of life--including your own.
-
S sebastian@social.itu.dk shared this topic
-
@gerrymcgovern
Can someone please explain to me how data centres will 'consume' so much water? There are plenty of energy intensive industries that have a heat dissipation need. If there's so much energy to dissipate put a fucking big heat exchanger out at sea and think about making them nursaries for fish/coral or something. It seems MAD to keep data centres where shifting the heat means single-use water. Use brine or molten salt or puréed billionaires in a closed loop heat exchanger near a heat sump where heat damage will be containable & acceptable.OR just say, "the price for that much AI is unsustainable. Sorry, but NO!"
@gregalotl "out at sea": the ocean life is extremely sensitive to the smallest temperature changes ... but your idea of pureed billionaires has good vibes.
@gerrymcgovern -
"AI is becoming something much larger. It is becoming one of the most important energy stories of the decade."
So says Big Oil
AI is an energy industry play.
"The scale of demand is extraordinary."
AI is a life destroyer. If you use AI, you are an active participant in the destruction of life--including your own.
@gerrymcgovern I would differ between AI and AI, please. These datacenters are mostly for the popular uses of LLMs, GPTs and Diffusion models. This mass crap is the destroyer.
But AI in the form of MLs (Machine learning) is a part of science. We wouldn't have modern climate science, if we had no ML which is part of AI! Would you really want to condemn scientists for doing this work?
-
@gerrymcgovern I would differ between AI and AI, please. These datacenters are mostly for the popular uses of LLMs, GPTs and Diffusion models. This mass crap is the destroyer.
But AI in the form of MLs (Machine learning) is a part of science. We wouldn't have modern climate science, if we had no ML which is part of AI! Would you really want to condemn scientists for doing this work?
@NatureMC @gerrymcgovern
I broadly agree. Generative LLMs Vs plain machine learning. -
@gerrymcgovern I mean it‘s possible we get some improvements in medical diagnostics for instance - but I guess it’s a long way until billions of people benefit from this.
@stephan324ppm We should differ between scientific machine learning and popular genAI.
Example: In our hospital they now have MRT scanners that use much less electricity (these machines use a lot). It is a machine learning (search Wikipedia) program = AI changing it for the good. But that has nothing to do with LLMs or GPTs.
@gerrymcgovern -
@gerrymcgovern I would differ between AI and AI, please. These datacenters are mostly for the popular uses of LLMs, GPTs and Diffusion models. This mass crap is the destroyer.
But AI in the form of MLs (Machine learning) is a part of science. We wouldn't have modern climate science, if we had no ML which is part of AI! Would you really want to condemn scientists for doing this work?
What you are referring to as being essential for modern (climate) science seems to be just software, algorithms, which now reversely are called "machine learning" and "AI" in order to make the failed "AI" project more convincing.
-
Imagine something that in 2020 didn't really exist but has so exploded onto our environment that in 2030 it will demand the water of 1.3 billion people and the electricity of 2 billion.
In a time when we have a global historic severe drought and when our civilization is about to collapse due to over-consumption of energy.
This is AI.
This is Big Tech
These are the tech bros. Here to burn it all down for greed and ambition.
Data centers could consume 9.3 Trillion Liters of Water by 2030
If treated as a country, data centers could rank sixth globally for electricity consumption by 2030.
This amount of electricity consumption carries an enormous carbon footprint – 189 million tonnes of CO2 equivalent, which only 3.2 billion tree seedlings grown over 10 years would be able to offset.
But the top tech bro from Microsoft says they only use as much as a restaurant
-
What you are referring to as being essential for modern (climate) science seems to be just software, algorithms, which now reversely are called "machine learning" and "AI" in order to make the failed "AI" project more convincing.
@sebastian I wrote my first article about AI research in the early 1980s. The word "artificial intelligence" is much older in science than in marketing. Yes, they worked with algorithms etc.
Yes, the marketing and the techbros claimed the older word for their crap. But how can we differ between the methods, if we just say "AI" for everything? If most people do that?
Not easy to find the language or reclaim words. But AI and ML is older than the big AI corporations.
@gerrymcgovern -
@sebastian I wrote my first article about AI research in the early 1980s. The word "artificial intelligence" is much older in science than in marketing. Yes, they worked with algorithms etc.
Yes, the marketing and the techbros claimed the older word for their crap. But how can we differ between the methods, if we just say "AI" for everything? If most people do that?
Not easy to find the language or reclaim words. But AI and ML is older than the big AI corporations.
@gerrymcgovern@NatureMC @gerrymcgovern yes, well aware of the history of "AI" (and ML) - a suggestion - if we stop calling general algorithms, deterministic and analytical software "AI", then we could avoid that trap of having to distinguish between good and bad "AI"
-
Data centers could consume 9.3 Trillion Liters of Water by 2030
If treated as a country, data centers could rank sixth globally for electricity consumption by 2030.
This amount of electricity consumption carries an enormous carbon footprint – 189 million tonnes of CO2 equivalent, which only 3.2 billion tree seedlings grown over 10 years would be able to offset.
But the top tech bro from Microsoft says they only use as much as a restaurant
@gerrymcgovern I'm telling you, they should be given fracking waste and told that they'd better treat it to drinkability or they won't get any water at all.
-
@NatureMC @gerrymcgovern yes, well aware of the history of "AI" (and ML) - a suggestion - if we stop calling general algorithms, deterministic and analytical software "AI", then we could avoid that trap of having to distinguish between good and bad "AI"
@sebastian The problem is: can you just stop to name it AI when it's officially part of it and a majority calls it like this? Will people understand your turn?
My way is that I try to name the systems instead of a blank "AI". But then people ask me what a LLM is. Many don't know it! (Even users!) And then I have to talk about all of it .. I need a wording. One can't decide about words, language lives from all people and takes time.
@gerrymcgovern -
J jwcph@helvede.net shared this topic
