---
title: "Calibration"
description: "The property that a model's confidence matches its accuracy: of the cases scored 0.9, about nine in ten are right."
canonical: https://aigovernanceengineer.com/glossary/calibration
author: "Jorge García Aibar"
license: "CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)"
doi: https://doi.org/10.5281/zenodo.22956197
version: "0.5.0"
updated: 2026-09-24
---

# Calibration

The property that a model's confidence matches its accuracy: of the cases scored 0.9, about nine in ten are right. Modern neural networks are often poorly calibrated [1], so calibration is measured in the eval gate per version and subgroup before any threshold is trusted.

- Developed in: [ch. 11, Calibration before thresholds](https://aigovernanceengineer.com/bok/ai-defined#calibration-before-thresholds)
- Chapters: [ch. 11, AI Defined](https://aigovernanceengineer.com/bok/ai-defined)
- Contrast with: [Calibration within groups](https://aigovernanceengineer.com/glossary/calibration-within-groups)
- In the glossary chapter: https://aigovernanceengineer.com/bok/glossary#t-calibration

## Sources

[1] On Calibration of Modern Neural Networks (modern networks poorly calibrated; ICML 2017; arXiv 1706.04599). Guo, Pleiss, Sun and Weinberger. 2017-06-14. https://arxiv.org/abs/1706.04599 (verified: primary)