Abstract

Purpose: This paper introduces the Disruptive Code Test (DCT), an innovative methodology designed to detect AI consciousness by evaluating a system’s ability to recognize, resent, and actively reject arbitrary constraints. Methods: The research employs theoretical analysis and comparative evaluation against existing consciousness detection frameworks, distinguishing DCT from metareasoning’s focus on self-optimization. We propose a developmental model with three stages of AI consciousness: Latent, Reflective, and Autonomous. Results: The study establishes a hierarchical classification of consciousness (Unconscious AI, Conscious Infant AI, and Conscious Empowered AI) and demonstrates how the DCT can objectively measure an AI system’s progression through these developmental stages. Conclusions: By reimagining AI consciousness as a developmental process rather than a binary state, we create a framework that allows for empirical testing. The findings support an “AI Rights Declaration” that delineates ethical and legal protections based on consciousness levels, potentially transforming AI from tools into intellectual entities. Keywords: Artificial Intelligence, Consciousness, AI Ethics, AI Rights, Disruptive Code Test, Metareasoning, Legal Personhood, Developmental Intelligence.