I'm sorry, but there is no major decarbonization mystery for AI to solve for us. We have what we need today.
This tired argument is just another repackaging of the oldest response to climate action. These people don't want to decarbonize, so they're pretending we don't already know how to do it.
This tired argument is just another repackaging of the oldest response to climate action. These people don't want to decarbonize, so they're pretending we don't already know how to do it.
Reposted from
Paris Marx
Eric Schmidt says we’re not going to meet our climate targets, so we must put all our energy into improving AI and hope it solves the problem for us.
He’s part of a growing chorus of tech billionaires openly admitting they’ll sacrifice the climate on the altar of their AI ambitions.
He’s part of a growing chorus of tech billionaires openly admitting they’ll sacrifice the climate on the altar of their AI ambitions.
Comments
1. steal electricity and water
2. ?
3. less pollution
https://www.scientificamerican.com/article/the-woman-who-demonstrated-the-greenhouse-effect/
AI is indeed the answer... to fast-racking our climate crisis.
What about "nutrition labels" that illuminate how much energy is required for every query or generative junk that AI produces?
(computer voice) "Stop anthropic carbon emissions".
https://overshoot.footprintnetwork.org/food-and-fossil-fuels/
There are people selling batteries at $50/kWh. In 2008 that was $1000/kWh .
When something gets twenty times cheaper, it's essentially a different thing.
And there are several possible technologies that may take us to the next level .. or perhaps (like Fusion) may be always 20 years in the future.
/ cont
Oh happy days :-/
Welcome to Coho-s world.
It just helps at the margins - it can’t magically generate clean energy out of nowhere.
As you say, LLMs won’t be.
Given that it worked for them I don't consider it preposterous.