Perhaps because "A.I." doesn't understand anything, it just makes plausible output based on its training data. LLM is a much better term, because intelligence

Perhaps because "A.I." doesn't understand anything, it just makes plausible output based on its training data. LLM is a much better term, because intelligence has connotations of understanding, whereas model does not.