I obviously meant this tweet as a joking but reading pmarca's and balaji's takes on AI safety it's clear they have absolutely no fear or conception that AI may outsmart them. It's just math it's just words it can just be turned off.