102
Energy companies told to recharge for AI datacenter surge
(www.theregister.com)
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 12 Oct 2024
102 points (100.0% liked)
Technology
58633 readers
3591 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
The study this cites has data centre (so not just AI but all internet stuff) rising to 300TWh by 2030. Two years ago the USA's power usage was 4000TWh a year. So in about 6 years time they estimate that data centres will be using about 8% of 2022's electricity usage, up from currently about 4%. An increase sure, but hardly one that's going to move electricity prices significantly.
If EV get out of their slump, they will devour every last electron that we have and we will need 10 times the current production.
I can't find the source but I remember an article thatdiscussed the rate solar energy is adopted. The researchers made lower and upper bound predictions, and what if all solar PV development stopped immediately. The worst case scenario based on availabile data suggested a three fold increase in solar PV electricity generation, the number used by the article you cited, to a best case scenario of solar PV increasing to the power of three, really big exponential growth. Now the optimistic model seemed a bit too optimistic for me, but it at least suggested that there is a lot more capacity to build out solar PV. If that capacity is realized or wasted was the biggest unknown factor in that study, which like duh. Still, I took it to mean that the future will probably be a little bit more optimistic than the most pessimistic projections. It's a small comfort.
That not how prices work in a shortage.
I didn't say anything about how prices work in a shortage, but I also sincerely doubt a 4% increase in 6 years (so 0.7% annually) is going to cause any shortages.
So, you have other information you are basing that on? Because the source is one of the top 3 consulting firms in the world.
I'm sure they will happily provide consultancy to the energy sector to avoid this issue, but I'm kinda surprised when random Lemmy accounts think they know better than a leading company like this.
Energy is a pretty tight market where supply and demand are tightly intertwined. So a big boom on the use side as AI is threatening has a good potential to outpace the capacity of the supply side to scale up.. thus creating scarcity, driving up prices.
Youve missread that article, it is saying rising demand generally may cause shortages, and that there is also predictions of growing demand fron datacentres, not that the later is the main cause of the former. I fact they suggest growth in electricty demand of a quarter in 2 years, vastly more than the 4% in 6 years growth in datacentre demand.