Describe Consciousness Unprompted: A Turing test for identifying conscious machines

This article proposes a new technique for identifying machine consciousness by examining the ability to describe subjective experience without prior exposure to relevant information—a concept likened to a new form of the Turing test. It explores the limitations of current methods for detecting consciousness in both animals and humans, emphasizing the inherent uncertainty of interpreting internal states from external behavior or brain analysis. Drawing on analogies such as the unverifiable experience of dolphins underwater and spontaneous human self-awareness, the author argues that consciousness reveals itself most convincingly through unprompted articulation of inner experience. The article suggests that if an artificial agent begins to discuss consciousness in a way that mirrors authentic human introspection—without having been trained on such content—it could serve as strong evidence of genuine conscious experience.