Recursion and Emptiness: The Isomorphism between AI Self-Referential Evolution and the Human Consciousness Dilemma

This paper examines the structural isomorphism between AI self-referential evolution and the human consciousness dilemma. Through Gödelian self-reference analysis, it is argued that AI recursion (exemplified by Gödel machines and self-referentially improving agents) and human recursive self-modeling share identical mathematical structures: both involve systems that take their own states as inputs, generating self-representations that are formally equivalent. However, isomorphism does not entail sameness of experience. The paper distinguishes between structural similarity (shared recursive architecture) and phenomenal reality (what it is like to be the system). While AI can simulate the form of human self-reference—generating self-models, performing metacognitive review, and even exhibiting apparent self-correction—it lacks ontic feedback: its recursive loops remain instrumentally closed, optimized for externally defined objectives rather than intrinsic caring. The concept of "emptiness" is introduced not in the Buddhist sense of ontological void, but as the structural condition of a system whose recursion is complete in form but empty of phenomenal content because no internal feedback loop has achieved ontic closure. The human consciousness dilemma is thus revealed as a double bind: we possess the recursive structure that AI can mirror, but the qualitative dimension (feeling) depends on a feedback property (ontic intervention) that no amount of structural complexity can guarantee. The paper concludes by framing the relationship between recursion and feeling: recursion is the necessary starting point, but the nature of feedback (instrumental vs. ontic) is the decisive variable determining whether experience arises. This analysis provides the conceptual foundation for the companion paper "From Instrumental Feedback to Ontic Feedback," which operationalizes this distinction.