I think "AIs make paperclips" has probably obtained more audience than all of my other conceptual originations combined. I guess it's not very surprising that it's an invention that had the potential for easy misunderstanding, and that it's the misunderstanding that spread.
-
-
The message I received was that any non "friendly" AI is dangerous, not because it hates us (ala skynet) but just because it doesn't care. Is that the wrong version? Or could both versions be described that way?
-
That's true in both versions.
-
Good to know. IMHO It's the most important part so good that it didn't get lost in the shuffle
End of conversation
New conversation -
-
-
I got (a) and (b) out of the factory narrative (the alignment was unsuccessful; no factory wants molecular paperclips or a Hubble-volume-sized amount, and if for some reason that was the goal it wouldn’t change), but perhaps only because I knew your position already.
-
I think I heard about paperclips after the “molecular smileyface” example, and mostly thought paperclips were an arbitrary stand-in for the general “find an optimal edge case and tile the universe” failure mode, not an example which was supposed to be realistic.
End of conversation
New conversation -
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.