Great, then we established that I (generally intelligent human), and LLMs, both can't perform the task in question without specific training. We still don't know if given that specific training we (a specific human or a specific LLM) would be able to perform it.