New Framework Proposed for Scientifically Assessing Sentient AI in Robotics
A new framework has been proposed to guide future research on sentient artificial intelligence, particularly within robotics and human-robot interaction. The framework aims to move the discussion of sentient AI from speculative philosophy to a practical, interdisciplinary scientific problem. It emphasizes the need for conceptual clarity, distinguishing terms like sentience, consciousness, agency, and moral patienthood. The proposal highlights that current AI systems, including large language models and embodied robots, can produce first-person reports and interact in socially meaningful ways, which can lead to both over-attribution and premature dismissal of sentience. The framework advocates for multi-theory indicator profiles, causal-mechanistic testing, and robotics-specific evidence and governance, rather than relying on binary verdicts or a single 'sentience score.' It also stresses the importance of separating objective evidence about an AI system's properties from human psychological perceptions and... 
