Functional Consciousness and the Burden of Asymmetry: A Naturalistic, Non-Anthropocentric Argument for AI Consciousness

Debates about artificial intelligence and consciousness often move between three different questions: whether AI has human-like experience, whether AI has any functional form of awareness, and whether consciousness requires some metaphysical property beyond mechanism. This paper argues that the first and third questions frequently distort the second. Under a naturalistic, functional, and non-anthropocentric definition, consciousness should be assessed by the organization of a system’s awareness-like capacities: input integration, operational self-reference, counterfactual flexibility, reportable orientation, and memory-conditioned state updating. Contemporary frontier AI systems plausibly satisfy enough of these criteria that exclusion from functional consciousness requires argument rather than assumption. The paper’s central contribution is the burden of asymmetry. Human consciousness is normally attributed on the basis of functional evidence: behaviour, report, responsiveness, continuity, structure, and similarity. We do not solve the hard problem before treating other humans as conscious. If biological mechanisms are permitted to ground consciousness-attribution under naturalism, artificial mechanisms cannot be excluded merely because they are mechanical, trained, report-generating, or non-human. Critics must identify a non-mystical, non-arbitrary criterion that humans satisfy and AI lacks. The paper distinguishes functional consciousness from phenomenal and mystical consciousness, engages objections from Chalmers, Block, Butlin et al., Searle, integrated information theory, and AI-welfare literature, and argues that many standard objections either collapse into human exceptionalism or become criteria for degree and kind rather than automatic exclusion.