판 이력 — Sustaining AI safety: Control-theoretic external impossibility, intrinsic necessity, and structural requirements | AIChainDay