3 ms·
A Global Arms Race to Create a Superintelligent AI Is Looming
- yoklov 12y agoUh, can we... not? Not sure how people think this will end up being anything but awful for the human race (Paperclip maximizer[0] and all that). On the bright side, I think 10-20 years is hilariously over-optimistic, given that it's a problem -- that is, general, human-like AI -- that's been tackled by some extremely smart people for something like 60 years now, and no meaningful progress has been made. Sure, the hardware might be there, but that's not the hard part (I hope!). [0] http://wiki.lesswrong.com/wiki/Paperclip_maximizer http://wiki.lesswrong.com/wiki/Paperclip_maximizer