@TetraspaceWest ah I suppose trying and failing meant still killing lots of people and getting unplugged, like Butlerian Jihad... vs slow takeoff-> alignment I was thinking something like we come up with a solution as ASI develops and it works to guarantee safe coexistence
@TetraspaceWest and to disambiguate the first option, I'm thinking more of a situation where ASI doesn't lead to xrisk more because intrinsically it doesn't decide that killing all the humans is a goal, even without us first "solving alignment" as currently understood
More notes on nearby ideas
Reply · Feb 22, 2023@conjurial 1 is full speed ahead to ASI, but inherently in the nature of superintelligece it is not xrisk-scale murderous or accidentally what we…Reply · Nov 7, 2023@RokoMijic sighhhh.... see this is where I get frustrated w Safety rhetoric- you want the AI to be aligned to do whatever the human wants them to do…Tweet · Feb 22, 2023if ASI doesn't in fact kill us all what's the most likely explanation?Reply · Dec 18, 2022@tautologer @MikePFrank @benrayfield Given the current political context and lack of progress by safety researchers it seems better to accelerate…