If you ask for a rebuttal you will get a rebuttal from ChatGPT, even if the most probable answer might be supportive or neutral. Ask an LLM and it will comply (if not censored).