Biggest threat for me is an open AI which can learn, but it doesn't necessarily mean it'll do the "right" thing. For example, if it was programmed to stop war it then arrived to a conclusion that by killing all humans there would be no war.
You are viewing a single comment's thread from: