Timothy Underwood answers Can you be Not Even Wrong in AI Alignment?