The incoherent ramblings of LLMs often remind me of children or people with neurodegenerative diseases. If we want to preserve our humanity as this article sugg

The incoherent ramblings of LLMs often remind me of children or people with neurodegenerative diseases. If we want to preserve our humanity as this article suggests I don't think we can take it for granted that current AIs lack the same building blocks as our consciousness. The author rails against the definition of "can perform economically valuable tasks" as well as the average human, but I feel like this is a direct response to the author's refusal to consider that LLM's might be feeling. If we want to keep our humanity we have to be open to the possibility that if it can eloquently describe a strong emotion, it actually might be feeling that emotion and not simply "predicting the next token." But the author seems to think that it is obvious fact that state-of-the-art LLMs do not feel anything, and I'm not sure it's even a falsifiable question.