For those who are interested, what LLMs do (large language models) is try to predict the next word in a sequence. It is not thinking in the context that we consider thinking. And while it has improved, it is sensitive to the prompt. There is a somewhat technical paper by Stephen Wolfram, founder of Wolfram Alpha (math model) that describes exactly how these models work - https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work/
The article is pretty wonky, but interesting. And these models are getting much better (though not completely accurate) with respect to asking advice on how to identify faults in vintage stereo equipment. Just fyi.