Structured adversarial testing of a model or agent to elicit failures (jailbreaks, injection, tool misuse) before an attacker does; treated here as an evidence-producing control.
- Developed in
- ch. 05, Pattern: Adversarial Red-Team Suite
- Chapters
- ch. 04, The Stack · ch. 05, Patterns · ch. 15, Deployment
- 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.
- 01 · Definition The three questions 1 mention
- 04 · The Stack Opening 3 mentions
- 05 · Patterns Pattern: Eval Gate in CI 5 mentions
- 06 · The Role Opening 4 mentions
- 07 · Maturity Model Observable criteria, by layer and level 1 mention
- 08 · Regulatory Map Opening 4 mentions
- 10 · Reading List Regulation and standards 5 mentions
- 12 · Governance Program Policies across the lifecycle 1 mention
- 13 · Risk Management The loop: identify, assess, treat, monitor 2 mentions
- 14 · Development The build as a chain of gates 1 mention
- 15 · Deployment Periodic assurance 1 mention
- 16 · Fairness & XAI Fairness and explainability in the stack 1 mention
- 18 · EU AI Act How to read this chapter 1 mention
- 19 · Privacy & AI Obligation to artefact map 1 mention
- 20 · Existing Law How to read this chapter 1 mention
Patterns that use this term
5 pattern pages use the term, most mentions first.
- Eval Gate in CI 1 mention
- Adversarial Red-Team Suite 1 mention
- AI Threat Model 1 mention
- Fairness Eval Suite 1 mention
- Claims Substantiation Gate 1 mention
Related terms
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 .