i think i'm wrong & lazy to use "AI will be like humans" as prior for convos abt AI emotions & self deception & goals & etc but i don't have a good mental model for possibility space & forces acting to shape AGI's traits
More notes on nearby ideas
Reply · Nov 28, 2023@a_musingcat hahaha we really don't want AI to be too human because then what's the point!?Tweet · Sep 6, 2022How would you use an LLM to build AGI with its own goals and motivation?Repost · Apr 2, 2022RT @vgr: I’m turning into an AI contrarian of sorts. I think AGI is ill-posed bullshit and embodiment and information boundaries matter a l…Reply · Nov 9, 2023@MindEnjoyer lol at this vs prev conversation abt the value of truth & AIs being able to self examine and know their own motivations...