ai is here
Aug 26, 2017 7:26:07 GMT -5
Post by surnshurn on Aug 26, 2017 7:26:07 GMT -5
Well, I was more getting into science fiction rather than the current logic for AI, haha.
ha. i probably give this topic a lot more consideration and thought than your average bear, so please pardon me if i get a bit nerdy about it.
the youtube channel "computerphile" based at Nottingham College in the UK has a really good youtube channel with some pretty interesting discussion pieces on the topic. the discussion there isn't so much on how it can work, but on how it does work and how it should work. beyond the elevator-talk stamp collecting robot i'd mentioned here, they also talk about things like the stop button scenario, where a robot behaves badly and is given a stop button with high priority to not disable to stop button, but then proceeds to eliminate any conceivable party that could actually press the stop button. in those videos they also mention 'terrible accident' in a way that makes me pretty sure they've seen something like that happen before. if you look closely there is a pretty interesting disagreement between the professors there, with some against developing ai and some for... i guess it's my doctor who.
as an aside: the stop-button scenario robot could make for a pretty good videogame plotline, where an AI creates an entire military-industrial complex to protect its stop button, which it absolutely cannot allow to be disabled, but for that reason cannot allow to be pressed (because that would increase the likelihood of the stop button being disabled)