COMPEL Specialization Stream · AITM-AAG
COMPEL Academy — Agentic AI Governance Associate
Professional certification for practitioners governing autonomous and agentic AI systems, covering delegation frameworks, guardrail design, observability, and human oversight patterns.
Profession title: Agentic AI Governance Associate
Audience: AI governance practitioners, platform engineers, and risk managers overseeing autonomous and agentic AI deployments.
Enroll in the AITM-AAG track
Registration, enablement, and the proctored assessment are delivered through compel.one. Seats open continuously.
Prerequisite chain
- AITF AI Transformation Foundations
- → AITM-AAG (this credential)
Learning outcomes
The learning journey is sequenced to cover each outcome below in order. Every article in the journey maps to at least one outcome.
- 1. Design delegation and autonomy frameworks for agentic AI.
- 2. Implement guardrails and runtime safety patterns.
- 3. Build observability and audit infrastructure for autonomous systems.
Body of Knowledge articles (18)
Module M1.6 (18 items)
- Article What Agentic AI IsM1.6-Art01
- Article Agent Architecture Patterns and InventoryM1.6-Art02
- Article Autonomy ClassificationM1.6-Art03
- Article Delegation, Authority Chains, and Legal ImplicationsM1.6-Art04
- Article Human Oversight Design Under EU AI Act Article 14M1.6-Art05
- Article Tool-Use Governance and Excessive AgencyM1.6-Art06
- Article Memory Governance and Poisoning DefenseM1.6-Art07
- Article Multi-Agent Systems and A2A ProtocolsM1.6-Art08
- Article Agentic Risk TaxonomyM1.6-Art09
- Article Agent Observability and AuditM1.6-Art10
- Article Kill-Switch, Containment, and Incident ResponseM1.6-Art11
- Article Regulatory Obligations for Agentic SystemsM1.6-Art12
- Article Cross-Organizational and Supply-Chain AgentsM1.6-Art13
- Article The Agent Governance PackM1.6-Art14
- Lab Lab — Autonomy Classification ExerciseM1.6-Art51
- Lab Lab — Human Oversight Regime Design for a Finance AgentM1.6-Art52
- Case Study Case Study — Moffatt v. Air Canada (2024 BCCRT 149) as an Agentic Governance FailureM1.6-Art61
- Template Template — Agent Governance CharterM1.6-Art71
Competencies demonstrated
- → Agentic AI delegation and autonomy frameworks
- → Guardrail design and runtime safety patterns
- → Observability and audit trails for autonomous systems
- → Human oversight escalation and intervention models
Exam blueprint summary
- Assessment
- Non-proctored assessment
- Passing score
- 70% passing score
- Portfolio
- Not required
- Renewal
- Every 24 months
- Recommended hours
- 22
- CE credits
- 22
Linked Core Mastery context
The Specialization Stream assumes AITF Foundations fluency. These Core Mastery resources are the recommended grounding before entering the AITM-AAG learning journey.
Formal credential definition
The machine-readable Open Badges 3.0 / W3C Verifiable Credential
definition for AITM-AAG is published at
/credential/aitm-agentic-governance
. HR platforms and AI citation engines can fetch the JSON-LD
document at
/credential/aitm-agentic-governance.json
.