GovernanceCore
Trustworthiness

Explainability

Working definition

Reviewed 30 July 2026

Explainability is a property of an AI system and its supporting information that enables relevant people to understand the factors, logic, or process associated with an output. The explanation needed depends on the audience, decision, risk, and purpose.

Context

Why it matters

Without usable explanations, people may be unable to oversee outputs, contest adverse decisions, diagnose failure, or determine whether a system is operating as intended.

Operating note

What this looks like in practice

  1. 01Define the audience and decision the explanation must support.
  2. 02Validate explanation fidelity and usefulness; visual polish is not evidence of truth.
  3. 03Pair technical explanations with process, data, limitations, and recourse information.

Sources & further research

Primary authority anchors the definition. Research links add conceptual or operational depth. External sources may update independently; always verify legal duties against the current official text.

  1. Official source

    The Language of Trustworthy AI: An In-Depth Glossary of Terms

    Synthesizes terminology for explainability, interpretability, transparency, and related concepts.

    NIST
    2023    Open source ↗
  2. Research paper

    “Why Should I Trust You?”: Explaining the Predictions of Any Classifier

    Introduces LIME and frames explanations as tools for judging individual predictions and model behavior.

    Ribeiro, Singh, and Guestrin
    2016    Open source ↗