Explainable AI focuses on techniques that make a model’s reasoning transparent and understandable. Methods include saliency maps, Shapley values, and interpretable surrogates. XAI is vital for trust, safety, and compliance, especially in sensitive applications.