“LLMs exhibit behaviour that is consistent with the outputs of mentalistic inference in humans”

Link. We are not magical.

“importance of systematic testing to ensure a non-superficial comparison between human and artificial intelligences”