Skip to content
Sign in

What is Explainability?

Also known as: XAI

Explainability, also known as Explainable AI (XAI), refers to methods that make an AI system's decisions and outputs understandable to humans.

It involves techniques that reveal how inputs lead to specific outputs, such as highlighting influential features or generating human-readable rules.

Key ideas include distinguishing between inherently interpretable models like decision trees and adding explanations to complex black-box models after training.

In ethics, it supports accountability by helping detect bias, ensure fairness, and meet regulatory requirements for transparency.

Example

A bank uses an AI model to approve loans; explainability shows an applicant that their low credit score and high debt-to-income ratio were the main reasons for denial.

Why it matters

It builds user trust and enables oversight in high-stakes applications like healthcare and justice, while helping organizations comply with emerging AI regulations.

Frequently asked questions

Interpretability means a model is simple enough to understand directly, while explainability adds explanations to more complex models.