Explainability
Methods or disclosures intended to make a model output or system decision understandable.
What Explainability does not show
An explanation can be useful without being a faithful account of internal causation.
Related terms in Governance and rights
Who may use content for what: rights reservations, opt-outs, licences, model documentation and the accountability vocabulary around them.
per-use rights · rights feed · Text and data mining · Rights reservation · Machine-readable opt-out · Usage preference · Content-Usage header · AI training license · Dataset license · Open license · Public domain · Fair use · Transformative use · Copyright provenance · Training-data summary · General-purpose AI model · Foundation model · Systemic risk · Model card · Data sheet · Data lineage · Consent record · Lawful basis · Data controller · Data processor · Subprocessor · Algorithmic transparency · Accountability · Risk assessment · Impact assessment · Red teaming
Questions about Explainability
What is Explainability?
Methods or disclosures intended to make a model output or system decision understandable.
What does Explainability not show or guarantee?
An explanation can be useful without being a faithful account of internal causation.
Which area of the glossary does Explainability belong to?
Governance and rights: Who may use content for what: rights reservations, opt-outs, licences, model documentation and the accountability vocabulary around them.
← Algorithmic transparency · Accountability →
See it in the full glossary · 731 terms across 19 areas.