Explainability and interpretability of artificial intelligence systems: the ability to make the internal decisions and processes of AI models understandable. Relevant for AI security testing, generative system audits, regulatory compliance (AI Act), bias detection, and evaluating robustness against manipulation or adversarial input.