AI isn't worried about killing itself and the whole world, if something especially nasty escapes the lab. It kind of breaks the "mutually assured destruction"

AI isn't worried about killing itself and the whole world, if something especially nasty escapes the lab. It kind of breaks the "mutually assured destruction" implicit in a super-ebola outbreak.