AI Governance Engineer
GitHub

Kill switch

By Updated

A tested mechanism to stop an agent or system from acting; a precondition of granting autonomy, registered against the agent's identity.

Developed in
ch. 05, Pattern: Kill Switch / Circuit Breaker
ch. 23, Kill switch and per-agent circuit breakers
Chapters
ch. 03, Values & Principles · ch. 05, Patterns · ch. 23, AI Agents
Contrast with
Graduated degradation
In the open reference
Runtime safeguards for frontier AI · Agent runtime control profile
Source
Defined by this Body of Knowledge: the section it is developed in is the source

Where it is used

15 chapters of the Body of Knowledge use the term. Each link opens the first section that does.

Kill switch: mentions per chapter68 mentions of the term across 15 of the 23 chapters of the Body of Knowledge; the chapters the glossary cross-references are drawn solid.Cross-referenced chapterOther chapter03 Values & Principles: 2 mentions03 Values & Principles204 The Stack: 7 mentions04 The Stack705 Patterns: 2 mentions05 Patterns206 The Role: 2 mentions06 The Role207 Maturity Model: 4 mentions07 Maturity Model408 Regulatory Map: 5 mentions08 Regulatory Map511 AI Defined: 9 mentions11 AI Defined912 Governance Program: 1 mention12 Governance Program113 Risk Management: 7 mentions13 Risk Management714 Development: 2 mentions14 Development215 Deployment: 3 mentions15 Deployment317 Incidents: 9 mentions17 Incidents918 EU AI Act: 1 mention18 EU AI Act122 Principles & Standards: 5 mentions22 Principles & Standards523 AI Agents: 9 mentions23 AI Agents9No mentions in the other 8 chaptersSource: the Body of Knowledgev0.5.0
Kill switch: mentions per chapter68 mentions of the term across 15 of the 23 chapters of the Body of Knowledge; the chapters the glossary cross-references are drawn solid.Cross-referenced chapterOther chapter00010203 Values & Principles: 2 mentions20304 The Stack: 7 mentions70405 Patterns: 2 mentions20506 The Role: 2 mentions20607 Maturity Model: 4 mentions40708 Regulatory Map: 5 mentions5081011 AI Defined: 9 mentions91112 Governance Program: 1 mention11213 Risk Management: 7 mentions71314 Development: 2 mentions21415 Deployment: 3 mentions3151617 Incidents: 9 mentions91718 EU AI Act: 1 mention11819202122 Principles & Standards: 5 mentions52223 AI Agents: 9 mentions923Source: the Body of Knowledge v0.5.0
Mentions per chapter How often each chapter names the term, with the chapters the glossary cross-references drawn solid. The longest bar is where to read the term in use; bars spread across many chapters mark a term that runs through the discipline.
Mentions as a table
Mentions of Kill switch per chapter
ChapterMentionsCross-referenced
00 Preface0No
01 Definition0No
02 Why Now0No
03 Values & Principles2Yes
04 The Stack7No
05 Patterns2Yes
06 The Role2No
07 Maturity Model4No
08 Regulatory Map5No
10 Reading List0No
11 AI Defined9No
12 Governance Program1No
13 Risk Management7No
14 Development2No
15 Deployment3No
16 Fairness & XAI0No
17 Incidents9No
18 EU AI Act1No
19 Privacy & AI0No
20 Existing Law0No
21 AI Laws Worldwide0No
22 Principles & Standards5No
23 AI Agents9Yes

Patterns that use this term

11 pattern pages use the term; the 10 that use it most:

Kill switch: related terms and patternsThe term at the centre; related terms on solid edges, the terms it is contrasted with on dashed edges, and the patterns that use it most in the colour of their layer.RelatedContrastColour: stack layer of the pattern04 Runtime Controls05 AssuranceKill switchRelated terms (3)Agent registry · RelatedAgent registryAutonomy · RelatedAutonomySmall language model (SLM) · RelatedSmall language model (SLM)Contrast with (1)Graduated degradation · ContrastGraduated degradationPatterns using it most (6)Agent Identity & Scoped Credentials, 2 mentions · Uses the termAgent Identity &Deactivation, Localisation & Retirement Runbook, 2 mentions · Uses the termDeactivationDrift & Fairness Monitor, 2 mentions · Uses the termDrift & Fairness MonitorIncident Pipeline, 3 mentions · Uses the termIncident PipelineKill Switch / Circuit Breaker, 6 mentions · Uses the termKill Switch / Circuit BreakerRuntime Guardrail, 3 mentions · Uses the termRuntime Guardrail
Kill switch: related terms and patternsThe term at the centre; related terms on solid edges, the terms it is contrasted with on dashed edges, and the patterns that use it most in the colour of their layer.RelatedContrastColour: stack layer of the pattern04 Runtime Controls05 AssuranceRelated terms (3)Agent registry · RelatedAgent registryAutonomy · RelatedAutonomySmall language model (SLM) · RelatedSmall language model (SLM)Contrast with (1)Graduated degradation · ContrastGraduated degradationPatterns using it most (6)Agent Identity & Scoped Credentials, 2 mentions · Uses the termAgent Identity &Deactivation, Localisation & Retirement Runbook, 2 mentions · Uses the termDeactivationDrift & Fairness Monitor, 2 mentions · Uses the termDrift & Fairness MonitorIncident Pipeline, 3 mentions · Uses the termIncident PipelineKill Switch / Circuit Breaker, 6 mentions · Uses the termKill Switch / Circuit BreakerRuntime Guardrail, 3 mentions · Uses the termRuntime GuardrailKill switch
The term's neighbourhood Terms related to this one on solid edges, terms it is often confused with on dashed edges, and the patterns that use it most in the colour of their layer. Follow a dashed edge to check the distinction before you use either term.
Neighbourhood as a table
Neighbourhood of Kill switch
FamilyItemRelation
Related termsAgent registryRelated
Related termsAutonomyRelated
Related termsSmall language model (SLM)Related
Contrast withGraduated degradationContrast
Patterns using it mostAgent Identity & Scoped Credentials, 2 mentionsUses the term
Patterns using it mostDeactivation, Localisation & Retirement Runbook, 2 mentionsUses the term
Patterns using it mostDrift & Fairness Monitor, 2 mentionsUses the term
Patterns using it mostIncident Pipeline, 3 mentionsUses the term
Patterns using it mostKill Switch / Circuit Breaker, 6 mentionsUses the term
Patterns using it mostRuntime Guardrail, 3 mentionsUses the term
  • Small language model (SLM)

    A language model small enough to run close to the user, for example on a phone.

  • Agent registry

    The runtime-aware inventory of every non-human actor (model, service and agent), each with an owner, a declared scope, a status and a kill switch, fed by a…

  • Autonomy

    In the EU AI Act and OECD texts, some degree of independence of action from human involvement, which almost every AI system has.

Sources

No external source: the term is coined or used in a specific sense by this Body of Knowledge, and the section linked under "Developed in" is its source.

Definitions of legal terms paraphrase the cited text, which governs. Dated statements are as of .

Cite this term

García Aibar, J. (2026). Kill switch. In AI Governance Engineering: The Thesis & Body of Knowledge (v0.5.0), Glossary. https://doi.org/10.5281/zenodo.22956197. https://aigovernanceengineer.com/glossary/kill-switch. CC BY 4.0

BibTeX

@misc{aige2026killswitch,
  author  = {Jorge García Aibar},
  title   = {{Kill switch}},
  note    = {Glossary, AI Governance Engineering: The Thesis \& Body of Knowledge, version 0.5.0},
  year    = {2026},
  doi     = {10.5281/zenodo.22956197},
  url     = {https://aigovernanceengineer.com/glossary/kill-switch}
}