Which principle requires AI systems to be understandable to non-technical stakeholders?
- Fairness
- Transparency
- Accountability
- Privacy
Answer: Transparency
Transparency (Explainable AI/XAI) ensures AI decisions can be interpreted by humans. Techniques: LIME, SHAP, attention visualization. Critical for trust, debugging, regulatory compliance.