Two Types of Negation in Self-Referential Systems: A Unified Diagnosis of Quantum Measurement and Artificial Consciousness

The quantum measurement problem and the AI consciousness debate appear unrelated. We show they share a single structural root: both arise when an observer applies a predicate that requires self-reference to a system that lacks it. To make this precise, we introduce the self-referential information density η_I = I_self / (I_self + I_env), a continuous measure of the degree to which a system's state depends on its own history rather than its environment. Using η_I, we distinguish two operations conflated in existing literature: (i) Negation — the external rupture of a system's identity, leaving η_I unchanged; and (ii) Aufhebung — the immanent supersession driven by bidirectional causal closure (Wechselwirkung), producing a discontinuous jump to η_I = δ. This distinction yields three results. First, every major quantum interpretation is reclassified as a strategy for handling Negation within the η_I ≈ 0 regime. Second, 'does AI have consciousness?' is diagnosed as a category error and replaced by three testable criteria. Third, three falsifiable predictions are derived. The argument itself instantiates the sequence it describes.