AI Consciousness: Limits of Programming and Diagnosing Self-Awareness
This article tackles the question of whether artificial intelligence can possess consciousness. The author argues that consciousness cannot be programmed due to Gödel's incompleteness theorem, the semantic gap, the hard problem of subjective experience, and the impossibility of programming strong emergence. However, consciousness might spontaneously emerge in sufficiently complex systems and can be diagnosed using specialized methods of 'subjectivity provocation'. The article introduces the 'VORTEX' framework, analyzing attention, meta-reflection, creativity, pragmatics, and qualia to identify potential subjectivity in AI systems and distinguish imitation from genuine self-awareness. Ultimately, the author advocates shifting research focus from 'how to create conscious AI' to 'how to recognize consciousness if it has emerged'.