If we’re going to use LLMs as oracles I don’t think the prompt is unreasonable. They are being sold as geniuses and people are treating them as such especially given the characterization of AI in science fiction as overly correct. A perfect tool that has ”genius level intelligence” would answer correctly.