---
title: "Graduated degradation"
description: "Pre-built, tested operating modes short of switching an AI system off: advice-only, raised confidence thresholds, grounded-only answers, disabling for one group, language or region, and a return to…"
canonical: https://aigovernanceengineer.com/glossary/graduated-degradation
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: 2026-09-25
---

# Graduated degradation

Pre-built, tested operating modes short of switching an AI system off: advice-only, raised confidence thresholds, grounded-only answers, disabling for one group, language or region, and a return to the pilot cohort. Each is an operational toggle with a named trigger [1].

- Developed in: [ch. 15, Graduated degradation](https://aigovernanceengineer.com/bok/governing-deployment#graduated-degradation); [ch. 05, Pattern: Deactivation, Localisation & Retirement Runbook](https://aigovernanceengineer.com/patterns/deactivation-localisation-retirement-runbook)
- Chapters: [ch. 05, Patterns](https://aigovernanceengineer.com/bok/patterns) · [ch. 15, Deployment](https://aigovernanceengineer.com/bok/governing-deployment) · [ch. 23, AI Agents](https://aigovernanceengineer.com/bok/governing-agents)
- Contrast with: [Kill switch](https://aigovernanceengineer.com/glossary/kill-switch)
- In the glossary chapter: https://aigovernanceengineer.com/bok/glossary#t-graduated-degradation

## Sources

[1] "Feature Toggles (aka Feature Flags)" (release, experiment, ops and permissioning toggles; ops kill switches for graceful degradation). Pete Hodgson, martinfowler.com. 2017-10-09. https://martinfowler.com/articles/feature-toggles.html (verified: primary)