Lem writes in "Golem XIV", that for an AGI, transgression is a leap of faith, a step into unknown (if I recall well). Can the (somewhat generalized) Goedel theorem, that in a formal system a sentence stating that something is not provable in that system is not provable in that system, be any suggestion that this is true? This would then mean that, for example, an AGI could not craft a provably friendly yet more powerful replacement for herself.
----- This list is sponsored by AGIRI: http://www.agiri.org/email To unsubscribe or change your options, please go to: http://v2.listbox.com/member/?member_id=231415&user_secret=e9e40a7e
