Most organizations have an AI explainability policy. Very few can answer the question a regulator or jury actually asks: why did this person, in this situation, receive this outcome? That's a different standard. Most organizations haven't built it.