AI Glossary Term
What does “XAI (Explainable AI)” mean?
Definition
Techniques and practices for making AI predictions more transparent and understandable.
How XAI (Explainable AI) works
Example
A hospital tool predicts a patient's risk of readmission and highlights which factors, such as recent admissions or medication count, contributed most. A doctor can judge whether those reasons make sense.
Why it matters
Explanations support trust, debugging, and accountability. They may be incomplete approximations.
Common questions
XAI vs explainability?
XAI is the field and its set of techniques; explainability is the property of being explainable.
Common techniques?
Feature-importance methods such as SHAP and LIME, saliency maps, and simpler surrogate models.
Are explanations reliable?
Not always. They can differ from what the model really does, so they should be validated.