As I understand it, ChatGPT is a predictive language model. It reflects human language through really good guesses based on machine learning algorithms munching on lots and lots of human-written content.

On the sentience/salience/consciousness scale, it’s closer to a guitar amplifier than it is to a conscious being.

AI is going to be a world-changing efficiency tool — but we seem to be quite a ways off from needing to wrestle with personhood ethics questions.