Explainability
Also known as: XAI · model explainability
Explainability is the ability to articulate why a model produced a given output or decision.
- Point to the features/evidence that drove it.
- Explain a decision in human terms.
- The difference between trusting a result and understanding it.
The practical bar: for high-stakes use, you should be able to explain the "why" to a non-expert. If you can't, you probably shouldn't be making that decision autonomously.