4 ms·
The author seems confused about what "alignment" means in the context of AI. Either that or they think our future robot overloads are the same thing as grandch
by TheDudeMan 4y ago
The author seems confused about what "alignment" means in the context of AI. Either that or they think our future robot overloads are the same thing as grandchildren (culturally different)?
- cwillu 4y agoSome history you may not be aware of: the lesswrong sequences were originally an extended discussion between Yudkowsky and Hanson, held via blog post on overcomingbias.org. Which isn't to say I disagree; if anything I find it a bit sad that he says things like “But even ignoring AI, our default future was never going to be aligned. […] These assumptions imply that our descendants will likely soon have very different cultures, including different values, beliefs, and behaviors. That is, they will be unaligned.” as not squelching humanity's choices was pretty explicitly an important part of alignment. See https://www.lesswrong.com/posts/HawFh7RvDM4RyoJ2d/three-worlds-collide-0-8 https://www.lesswrong.com/posts/HawFh7RvDM4RyoJ2d/three-worl..., which was originally posted on overcomingbias (see https://web.archive.org/web/20090608073156/http://robinhanson.typepad.com/overcomingbias/2009/01/three-worlds-collide.html https://web.archive.org/web/20090608073156/http://robinhanso... ) many years ago now.
- TheDudeMan 4y agoI know the history. Lol, I didn't even notice that Robin was the author.