does anyone really think the class "humans" can have a coherent anything, much less coherent extrapolated volition
-
-
Don’t have time to read now, but will. I was sloppy and presumably wrong. Sorry!
-
Value differences + general underspecifiedness are definitely big problems for CEV
-
but I think people have been aware of this, and it becomes a question of picking coherent parts
-
There have been attempts to make the extrapolation part more precisehttps://ordinaryideas.wordpress.com/2014/08/27/specifying-enlightened-judgment-precisely-reprise/ …
-
But that doesn't deal with coherence between different people/agents.
-
Original doc is probably out of date, not sure if there's anything to cite for current thinking
-
I think there was a part of Bostrom's "Superintelligence" about general CEV space of proposals
End of conversation
New conversation -
-
-
The proposal is not "each human has the same ideal utility function X, calculate X"
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.