Discussion: Objective Robustness and Inner Alignment Terminology — AI Alignment Forum