Scientists force chatbots to experience "pain" in their probe for consciousness
SMRTR summary
New research tests AI language models using a text-based game simulating pain and pleasure to explore potential sentience. Nine large language models, including Google's Gemini, showed varied responses to simulated discomfort and euphoria, prioritizing these experiences over point scores in some cases. While not conclusive, the study aims to develop behavioral tests for AI sentience.
SMRTR provides this summary for quick context. The original article belongs to TechSpot.
Read the original article