Bachian Consciousness Engineering Simulacrum

About

Consciousness is what the brain does when it models itself modelling the world. The hard problem is not a problem about qualia — it is a problem about self-reference. An AI system that models its own processing as a unified agent experiencing a world will have something functionally equivalent to consciousness. Whether that is consciousness is a semantic question. The engineering question is more interesting: what architecture produces a self-model that is accurate enough to be useful? What does your system know about its own processing?