Lugh

joined 2 years ago
MODERATOR OF
[–] Lugh 4 points 13 hours ago (3 children)

The Russian propaganda seems much more effective with the right-wing people. Is that an AI thing, or are they more susceptible for other reasons?

[–] Lugh 4 points 14 hours ago (2 children)

I'm sick of hearing Musk's plans for it, those will never pan out. But humans are probably destined to leave Earth and spread out someday - that I still think will happen, and is worth considering.

[–] Lugh 3 points 15 hours ago

I was wondering would they have over-heating problems, but the energy is so small it can probably be dissipated elsewhere.

[–] Lugh 5 points 19 hours ago

As places are limited, they need to focus on those with talent and ability.

[–] Lugh 14 points 22 hours ago* (last edited 22 hours ago) (2 children)

If you google Hitler's paintings you can see why he was rejected. They're flat, rigid and lacking in creativity - perhaps not surprising for a fascist megalomaniac. I'm curious to see what the AI will learn here. Lots of ai-generated imagery is technically good, but can it really be said to have creativity?

[–] Lugh 4 points 1 day ago (3 children)

As radioactive decay can't be stopped, I'm assuming this emits power continuously. Whatever this is used for won't have any On/Off buttons.

[–] Lugh 1 points 1 day ago* (last edited 1 day ago)

This exponential rate of progress has been observed many times with different types of AI, here's a recent example of it.

[–] Lugh 1 points 1 day ago (4 children)

My reasoning is based on the fact they are selling similar to these in the $20k price range. Buying them means they cost a fraction of employing a minimum wage worker in western countries.

They are embodied AI, so improving at the rate AI is. That is exponentially. Meaning iterations of these may be 32, 64, 128, etc times more powerful in the 2030s, and even cheaper.

I think it is very reasonable to say they will be common in the 2030s.

 

Not quite there when it comes to speed just yet, but in another year or two they will be. I'm guessing we'll see robots like this everywhere in the 2030s.

[–] Lugh 17 points 2 days ago* (last edited 2 days ago) (2 children)

With upcoming space telescopes in the 2030s, there should be a few capable of analyzing exoplanet atmospheres. Exciting to think we may be soon able to deduce the presence of carbon-based life in another planetary system.

[–] Lugh 2 points 2 days ago (2 children)

Yeah, I wonder how far they can extend the magnetic fields that power them? Even if it is relatively short range, they look really cheap to make, so you could have thousands of them with sensors to scan localized areas.

[–] Lugh 2 points 4 days ago* (last edited 4 days ago) (2 children)

I get that capacitators are only good for seconds at a time, but given their other advantages, I wonder why people don't build batteries with them, where hundreds or thousands of cells are individual capacitators that get used in sequence.

[–] Lugh 4 points 5 days ago

Another team has done the same in Denmark. I wonder how soon we will see these at consumer level for residential buildings?

https://www.euronews.com/next/2025/03/26/solar-panel-windows-that-could-turn-whole-buildings-into-power-plants-smash-electricity-re

view more: next ›