Oversight
AIGP glossary · Last reviewed: · By Victor Humenhuk (AIGP certified)
Oversight - Supervision arrangements, human or institutional, that monitor, review and can intervene.
In the AIGP body of knowledge, Oversight comes up under Module 8: AI Governance Vocabulary.
Oversight in context
- The [[Seven ethical issues]] responsible AI must address - lawfulness, safety, bias protection, transparency, choice, human intervention, security - and the [[Three foundational controls]] that mitigate ethical risk: ethical principles, an [[Oversight body]], and policies & procedures. (Seven ethical issues and three foundational controls)
- Policies must create oversight across ==nine areas== of the AI life cycle. (Life cycle policies and the use case assessment)
- The [[3×3 harms matrix]] multiplies severity × probability for a score, tolerances vary by organisation, and NIST calls for ==senior independent oversight== that declares risk tolerances. (Risk assessment mechanics)
Where Oversight is covered in the AIGP study notes
Related terms
- AI governance
- Accountability
- AI assurance
- AI audit
- Transparency
- Explainability
- Interpretability
- Contestability
- Conformity assessment
- Impact assessment
Test yourself on Oversight
Recognising a definition is not the same as applying it in an exam scenario. Work through the free AIGP practice questions, or read the full AIGP study guide - every study note is free.