Comment by trevyn

So how do superintelligence-AI-loving rationalists approach the notions that brains are quite efficient inference machines, silicon is a poor inferential substrate, and we don't know what the next approach for a substrate will be? That it'll work itself out eventually?