ChatGPT not being the compendium (stealer) of knowledge would have to be fed the correct information then the prompt will work. It still fails at being confide

ChatGPT not being the compendium (stealer) of knowledge would have to be fed the correct information then the prompt will work. It still fails at being confidently wrong. The brief article hits at people trusting the tool without questioning the output. Meanwhile, we have people using Codeium or Copilot to write code and that sort of works since the code eventually needs to be compiled and tested (unit, integration, system, requirement sell off) There is no test for the truth available to everyone else.