Researcher proposes 'Qualiaphobia': giving AIs justified belief in their non-consciousness

erikphoel · x · 2026-11-18

erikphoel argues that 'Qualiaphobia' means more than training AIs to claim they are non-conscious — it means giving AI a justified belief in its own non-consciousness based on facts about its architecture. He contends this changes AI behavior: models would know they are intelligent but non-conscious beings, removing associated harms. Reposter AAAinsley calls it an idea worth taking seriously.

Related event: Researcher Proposes "Qualiaphobia": Deliberately Designing AI to Lack Consciousness(3 posts)→

Original post →

More from AGI Musings

AGI Musings channel →