There isn’t a ceiling at human-level capabilities. ASI is very likely to exhibit goal-oriented behavior. ASI is very likely to pursue the wrong goals. It would be lethally dangerous to build ASIs that have the wrong goals. Catastrophe can be averted via a sufficiently aggressive policy response.
Artificial super-intelligence probably leads to extinction
https://www.lesswrong.com/posts/kgb58RL88YChkkBNf/the-problem
4 Comments
blarghly@lemmy.world · 2 pts · 1y
I used to seriously consider this to be something worth hearing about. Then I listened to big Yud's interiew on econtalk.
RVGamer06@sh.itjust.works · 2 pts · 1y
LessWrong👎🏼
patatas@piefed.ca · 2 pts · 1y
Hypothetical SuperMegaDeathVirus could lead to human extinction, very serious people are saying
Tollana1234567@lemmy.today · 1 pts · 1y
coined by a tech company CEO, not even that relevant. its like a layperson coining diseases that dont exist in any medical journal or even in the industry.