Kaj_Sotala comments on London meetup: “The Friendly AI Problem”