1. Safety breaks (ref. Azimov's laws of robotics)
  2. Goals and motivations: see ‣. Goals and motivations are not a consequence of intelligence and will not appear on their own.
  3. A general-purpose learning system (equivalent of the neocortex).

Additionally, in ‣, Pearl poses that causal explanations should become the cornerstone of machine intelligence (see Causal explanations, not dry facts, make up the bulk of human knowledge).


A Thousand Brains