ModelRefs / Counterfactual Explanation — AI Glossary
Counterfactual Explanation — AI Glossary
An explanation of a model's decision stating what minimal change to the input would change the model's output.
Overview
Counterfactual explanations (Wachter et al. 2017) answer: 'What would need to change for this loan to be approved?' They are actionable, user-understandable, and privacy-preserving (reference no training data). Required by GDPR Article 22 for automated decisions. Generated by optimization: find nearest input with different prediction.
Reference details
| Topic | safety |
|---|---|
| Also known as | counterfactual, what-if explanation |
| Last reviewed | 2026-06-24 |
Related terms
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Counterfactual Explanation — AI Glossary.
Frequently asked questions
What is Counterfactual Explanation?
An explanation of a model's decision stating what minimal change to the input would change the model's output.
Is Counterfactual Explanation the same as counterfactual?
Yes — counterfactual, what-if explanation are common aliases for Counterfactual Explanation.
What concepts are related to Counterfactual Explanation?
Closely related concepts include xai, lime, shap.