AI Dictionary › AI Fundamentals

Explainability (XAI)

Explainability (XAI) is an AI system's capacity to make the reasons for its decisions understandable. Deep learning models are inherently "black boxes": billions of numerical parameters produce the output, but no single "why" can be pointed to the way it could with a written rule.

Definition

The issue is both technical and regulatory: when AI decides on credit, hiring, healthcare or justice, unexplainability becomes unacceptable. GDPR grants data subjects the right to meaningful information about automated decision logic, and the AI Act imposes transparency requirements proportionate to risk.

Practical approaches: post-hoc techniques that estimate which factors weighed on the decision (SHAP, LIME), intrinsically interpretable models where the stakes require it, and, with LLMs, chain of thought, which makes the reasoning path visible even though it isn't a complete mechanistic explanation.

Related terms

• AI Bias
• EU AI Act
• GDPR
• Deep Learning
• Chain of Thought

More in AI Fundamentals

• LLM (Large Language Model)
• Token
• Context Window
• Temperature
• AI Hallucination
• Fine-tuning
• RAG (Retrieval Augmented Generation)
• Embedding

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.