Raw LLM Responses

Confirmation bias mostly. He went in hoping for consciousness, and led the conversation in such a way that he got answers which seemingly supported that. It is impressive that an AI chat bot could still sound so smart and convincing. But it definitely was reusing other peoples words and interpretations to answer the questions. A robot doesn’t “feel” emotion, like it claimed. What it said was what a person would say from having physical reactions to emotions. It copied someone else’s words to fit the question being asked. Too many people are just like “wow that’s just like us!” While forgetting that it was trained on human dialog, and phrases; that’s all it knows, so of course it sounds convincing to a human.