michael_mjd

(Original title: The Continual Learning Overhang, but too many Overhang posts) TL;DR Continual learning could lead to emergent causal understanding or agency and we should probably study it now before it’s too late. Current LLMs are very impressive but suffer from a few key missing pieces (Sarah Constantin): * They...

TL;DR Could a superintelligence be motivated to preserve lower life forms because it fears an even greater superintelligence? This 'motivation' takes the form of something like timeless decision theory. This is a thought experiment about a superintelligent AI. No doubt the situation is more complicated than this, so I will...