Clever! It will be interesting to see if OpenAI eventually programs around these sort of "hacks" or even ChatGPT will simple learn to recognize them. TY.