From Science Daily.
"The people who made LLMs call it "hallucinating" when they make things up; although Chemero says, "it would be better to call it 'bullsh*tting,'" because LLMs just make sentences by repeatedly adding the most statistically likely next word -- and they don't know or care...