Assorted links October 19, 2014 by Luke Hsu, “Super-intelligent humans are coming.” Winikoff, “Assurance of agent systems: what role should formal verification play?“ New Friendly AI open problem description: “Corrigibility.”