AI Governance Engineer
GitHub

Threat model (AI)

By Updated

A versioned record of what can go wrong with an AI system and what is done about it: data flows and trust boundaries, threats per element from STRIDE and AI-specific catalogues, a decision on each, and the test that proves each mitigation. It answers the four threat-modelling questions, ending with whether the job was done well enough 1.

Developed in
ch. 15, Threat modelling the deployed system
ch. 05, Pattern: AI Threat Model
Chapters
ch. 05, Patterns · ch. 14, Development · ch. 15, Deployment · ch. 23, AI Agents
Contrast with
Red teaming
Source
1 numbered reference, listed below

Where it is used

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

Threat model (AI): mentions per chapter32 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 chapter00 Preface: 1 mention00 Preface102 Why Now: 2 mentions02 Why Now203 Values & Principles: 2 mentions03 Values & Principles205 Patterns: 1 mention05 Patterns106 The Role: 1 mention06 The Role108 Regulatory Map: 4 mentions08 Regulatory Map410 Reading List: 2 mentions10 Reading List211 AI Defined: 2 mentions11 AI Defined212 Governance Program: 6 mentions12 Governance Program613 Risk Management: 2 mentions13 Risk Management214 Development: 1 mention14 Development115 Deployment: 1 mention15 Deployment119 Privacy & AI: 1 mention19 Privacy & AI122 Principles & Standards: 3 mentions22 Principles & Standards323 AI Agents: 3 mentions23 AI Agents3No mentions in the other 8 chaptersSource: the Body of Knowledgev0.5.0
Threat model (AI): mentions per chapter32 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 chapter00 Preface: 1 mention1000102 Why Now: 2 mentions20203 Values & Principles: 2 mentions2030405 Patterns: 1 mention10506 The Role: 1 mention1060708 Regulatory Map: 4 mentions40810 Reading List: 2 mentions21011 AI Defined: 2 mentions21112 Governance Program: 6 mentions61213 Risk Management: 2 mentions21314 Development: 1 mention11415 Deployment: 1 mention11516171819 Privacy & AI: 1 mention119202122 Principles & Standards: 3 mentions32223 AI Agents: 3 mentions323Source: 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 Threat model (AI) per chapter
ChapterMentionsCross-referenced
00 Preface1No
01 Definition0No
02 Why Now2No
03 Values & Principles2No
04 The Stack0No
05 Patterns1Yes
06 The Role1No
07 Maturity Model0No
08 Regulatory Map4No
10 Reading List2No
11 AI Defined2No
12 Governance Program6No
13 Risk Management2No
14 Development1Yes
15 Deployment1Yes
16 Fairness & XAI0No
17 Incidents0No
18 EU AI Act0No
19 Privacy & AI1No
20 Existing Law0No
21 AI Laws Worldwide0No
22 Principles & Standards3No
23 AI Agents3Yes

Patterns that use this term

5 pattern pages use the term, most mentions first.

Threat model (AI): 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 pattern01 Govern-as-Code02 Inventory03 EvalsThreat model (AI)Related terms (4)AI system · RelatedAI systemATLAS · RelatedATLASDual use · RelatedDual useSTRIDE · RelatedSTRIDEContrast with (1)Red teaming · ContrastRed teamingPatterns using it (5)AI Threat Model, 6 mentions · Uses the termAI Threat ModelAdversarial Red-Team Suite, 2 mentions · Uses the termAdversarial Red-Team SuiteDataset Admission Gate, 1 mention · Uses the termDataset Admission GateModel Artefact Integrity, 1 mention · Uses the termModel Artefact IntegrityUse-Case Intake & Risk Tiering, 1 mention · Uses the termUse-Case Intake & Risk Tiering
Threat model (AI): 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 pattern01 Govern-as-Code02 Inventory03 EvalsRelated terms (4)AI system · RelatedAI systemATLAS · RelatedATLASDual use · RelatedDual useSTRIDE · RelatedSTRIDEContrast with (1)Red teaming · ContrastRed teamingPatterns using it (5)AI Threat Model, 6 mentions · Uses the termAI Threat ModelAdversarial Red-Team Suite, 2 mentions · Uses the termAdversarial Red-Team SuiteDataset Admission Gate, 1 mention · Uses the termDataset Admission GateModel Artefact Integrity, 1 mention · Uses the termModel Artefact IntegrityUse-Case Intake & Risk Tiering, 1 mention · Uses the termUse-Case Intake & Risk TieringThreat model(AI)
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 Threat model (AI)
FamilyItemRelation
Related termsAI systemRelated
Related termsATLASRelated
Related termsDual useRelated
Related termsSTRIDERelated
Contrast withRed teamingContrast
Patterns using itAI Threat Model, 6 mentionsUses the term
Patterns using itAdversarial Red-Team Suite, 2 mentionsUses the term
Patterns using itDataset Admission Gate, 1 mentionUses the term
Patterns using itModel Artefact Integrity, 1 mentionUses the term
Patterns using itUse-Case Intake & Risk Tiering, 1 mentionUses the term
  • ATLAS

    MITRE's Adversarial Threat Landscape for Artificial-Intelligence Systems, a knowledge base of adversary tactics and techniques against AI, including…

  • STRIDE

    A threat-classification checklist from Microsoft's Security Development Lifecycle: spoofing, tampering, repudiation, information disclosure, denial of service…

  • AI system

    For governance, the object the AI definition brings into scope.

  • Dual use

    The capacity of the same AI capability to serve harmful ends as well as legitimate ones, for example a toxicity model inverted to propose toxic molecules.

Sources

  1. [1] Threat Modeling Manifesto (threat modelling as analysing representations of a system to highlight concerns about security and privacy characteristics; four key questions). Threat Modeling Manifesto working group. n.d. (accessed 2026-09-25). https://www.threatmodelingmanifesto.org/ (verified: primary)

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

Cite this term

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

BibTeX

@misc{aige2026threatmodelai,
  author  = {Jorge García Aibar},
  title   = {{Threat model (AI)}},
  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/threat-model-ai}
}