“Sometimes it seems as though each new step towards AI, rather than producing something which everyone agrees is real intelligence, merely reveals what real intelligence is not.” – Douglas Hofstadter
Douglas Hofstadter is an intellectual outlier, an iconoclast and a pariah to conventional wisdom on the topic of AI. These are all very good reasons to listen to him. As Eric Weinstein may have suggested in the excellent podcast published here recently, institutional networks (as a reflection of the aggregate self-interest of those entities who inhabit them) are predominantly concerned with the self-propagation and continuing tenure of their own organisational system(s).
Academic organisations are themselves self-replicating information systems, selecting for recombinatory continuity through minimising internal and external uncertainty or (systemic) entropy. Conventional platitudes and “safe” research agendas actively inhibit the free flow of ideas that discovery and authentic intellectual creativity or intelligence require. There is in this a logical clue as to the cause of an ever-receding vanishing-point of AGI. If Information Theory tells us that novelty is entropy and organisational, sociological and (associated) communications systems actively dampen such uncertainty, cathartic insights into AGI are very likely effectively obscured by an emergently complex psychological, cultural and epistemological blindspot.