On GPT 3.5 Q: "What is the seventy fourth element of the periodic table?" A: "The seventy-fourth element of the periodic table is Rhenium..." But this is rea

On GPT 3.5 Q: "What is the seventy fourth element of the periodic table?" A: "The seventy-fourth element of the periodic table is Rhenium..." But this is really shooting fish in a barrel. Given the way LLMs work why would you expect them to provide factually correct text completion?