The Internal Model Principle: A Straightforward Explanation — AI Alignment Forum