Artificial intelligence: the problem of ontological distinction between machine consciousness and its functional imitation

Abstract

The subject of the study is the ontological foundations for distinguishing between machine consciousness and the functional imitation of conscious activity in modern artificial intelligence systems. The analysis focuses on differentiating between functional intelligence, artificial agency, and the possible subjective existence of an artificial system. The study examines whether the reproduction of speech, memory, learning, adaptation, self correction, self description, and autonomous action is sufficient to recognize the presence of consciousness. Special attention is paid to the distinction between objectively measurable computational processes and states that may be given to the system itself as its own experience. The significance of systemic processual integrity, temporal continuity, an individualized history of interaction, a stable attitude towards the world, autonomous normativity, and the causal efficacy of the system’s representations of its own states is examined. The subject of the study thus encompasses the functional characteristics of intellectual behavior and the conditions for the possible emergence of non biological subjectivity. Method, research methodology. The methodological basis of the study consists of conceptual analysis, ontological reconstruction, the comparative method, and criticism of the functional reduction of consciousness, which make it possible to compare the observed properties of artificial systems with the conditions for the existence of subjective experience. The use of these methods ensures a consistent differentiation between functional intelligence, artificial agency, and machine consciousness. The scientific novelty of the study lies in the substantiation of the principle of ontological insufficiency of functional equivalence, according to which the coincidence of functions and the results of activity doesn’t mean the identity of the ways of existence of a human and an artificial system. A three level model is proposed, distinguishing between functional intelligence, artificial agency, and machine consciousness, as well as a system of interrelated criteria: systemic processual integrity, temporal continuity, individualized history of interaction, a stable attitude towards the world, autonomous normativity, causal effectiveness of the system’s representations of its own states, and the presence of its own subjective experience. It’s concluded that modern artificial intelligence systems demonstrate developed functional intelligence and certain signs of agency, but there are insufficient grounds to recognize that they possess independent subjective experience. Therefore, machine consciousness should be considered an open ontological possibility that requires stricter criteria than the external persuasiveness of behavior.