One of the ironies of current generative AIs and LLMs - they are creative, but need human supervision to watch for simple logical errors. Just the reverse of conventional way of looking at humans vs machines.