Comment on A Vaccine for Anthropomorphism of AIComments−waffletower1yWould also be helpful if we, and the author does not here, avoided use of inappropriately anthropomorphic concepts to describe LLM output, such as "hallucination". We already have a more appropriate word available, "confabulation".
Comments
Would also be helpful if we, and the author does not here, avoided use of inappropriately anthropomorphic concepts to describe LLM output, such as "hallucination". We already have a more appropriate word available, "confabulation".