Microsoft Needs So Much Power to Train AI That It's Considering Small Nuclear Reactors
Microsoft Needs So Much Power to Train AI That It's Considering Small Nuclear Reactors
Microsoft is betting on small modular nuclear reactors to grow its energy infrastructure for training AI, a highly ambitious plan.
You're viewing a single thread.
AI needs that's much power?
Fuck you, ditch it like a Zune and make some more video games.
Or make some more Zunes.
I miss my Zune. Oddly enough, every person I've met who had a Zune had it stolen, including me.
And they only ever made one.
Mine was never stolen, to break your streak. I had one of the little 4GB ones.
Technically he didn't meet you so the streak is still on.
So you're saying I need to stalk them down and introduce myself? Seems like a lot of work to avoid being incorrect, but since this is the internet I am obliged to do what I must to be correct.
More Zunes, please. My Zune 30 has dead pixels and the battery is on its way out.
Damn thing lasted longer than my marriage and an 8 year relationship after that.
The power consumption is factored into the cost of AI. It's still profitable after that, or they wouldn't be doing it.
It's the biggest buzzword right now, it doesn't matter if it's profitable. I doubt most uses are directly profitable right now. It'd more of a FOMO situation - if we don't use AI, we're OBSOLETE! AHHH!
If it turns out not to be profitable in the long run then people will stop.
Should we never even experiment?
That's basically exactly my point. Seems we agree lol. I was just poking fun at the fact that it feels like just about every large tech company is doing it, just like when the metaverse was all the rage... Or crypto, or...
Or Windows Phones 📱
Windows phones were nice, except Microsoft made them even more locked-down than iPhones...
Here's a nice video of a guy training an AI to do a relatively simple task (driving a Trackmania trac) with a very limited amount of inputs with low variability, 2-3 outputs and very hardset restraints.
Compared to what he does, a rather narrow defined re-enforcement training scheme, Microsofts AI takes many more inputs and has many more outputs and all the inputs are highly variable (massive amounts of data like dictionaries, images, movies, entire texts, speech, etc compared to a handful of parameters with values from -1 to 1) and also is a mix between re-enforcement, supervised and unsupervised training. With different subnetworks trained for different things eventually working together to do the master task they have in mind.
https://www.youtube.com/watch?v=Dw3BZ6O_8LY
What is shown in the video is what you'd do for a tiny subsystem of the AI Microsoft, Google, Apple and the like develop.
Kinda like if you watched a video about "this is what it takes to make the bolt that keeps your wheels on your car" you'd only have seen a fraction of what it takes to make the whole car.
Here is an alternative Piped link(s):
https://www.youtube.com/watch?v=Dw3BZ6O_8LY
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.