Model validation
What does model validation cover in AI governance?
2 competencies are filed under it: Model risk management and independent challenge, AI evaluation and testing design. Each has its own page saying what it means in practice.
The competencies filed under it
- Model risk management and independent challenge
Classifies models by tier, sets validation requirements, challenges data, methodology and performance evidence, and reports aggregate exposure.
Risk and Assurance
- AI evaluation and testing design
Designs tests for factuality, robustness, fairness, safety and abuse resistance with rubrics, baselines and thresholds, and says what a score misses.
Technical Evaluation
Roles that ask for it
- AI Model Validator2 of 2 competencies
- Model Risk Manager2 of 2 competencies
- AI Engineer, production systems edition1 of 2 competency
- AI Evaluation Specialist1 of 2 competency
- AI Governance Engineer1 of 2 competency
- AI Product Manager, responsible product edition1 of 2 competency
- Ethical AI Specialist1 of 2 competency
- Agentic AI Engineer, orchestration edition1 of 2 competency
- AI Security Architect1 of 2 competency
- Physical AI Safety Engineer, autonomous systems edition1 of 2 competency
- Robotics AI Engineer, embodied systems edition1 of 2 competency
- Chief Audit Executive, AI audit edition1 of 2 competency
- Chief Compliance Officer, AI compliance edition1 of 2 competency
- Chief Risk Officer, AI risk edition1 of 2 competency
- Chief Technology Officer, AI engineering edition1 of 2 competency
Where it is taught and graded
32 graded topics, each passed by explaining it back. The first module of every program is free with a free account.
- Module 2: Capability and Autonomy (1)
- Module 5: Technical Controls and Threat Modeling (1)
- Module 6: Testing and Red-Teaming (4)
- Module 11: The Capstone: Deploy and Defend (1)
- Module 1: Integrated Governance, Risk and Compliance and Enterprise Context (1)
- Module 4: Enterprise Risk Architecture and Risk Appetite (2)
- Module 6: Quantitative Risk and Decision Science (1)
- Module 12: Third-Party, Supply-Chain, and Concentration Risk (1)
- Module 17: AI Governance, AI Risk, AI Security, and AI Assurance (1)
- Module 18: Governance, Risk and Compliance Technology, Automation, and Evidence Engineering (1)
- Module 3: Failure Analysis, Learning From How Robots Break (1)
- Module 6: Human Factors and Human-Robot Interaction (1)
- Module 7: Research, Benchmarking, and Staying at the Frontier (2)
- Module 10: Responsible Physical AI in Practice (2)
- Module 4: Evaluation and Trust (3)
- Module 10: Evidence Engineering (1)
- Module 4: Practical AI Workflow Design and Prompt Engineering (1)
- Module 6: AI in the Workplace and Team Leadership (1)
- Module 13: Bonus: AI for Educators (1)
- Module 3: Business and Industry Representation (1)
- Module 6: Balancing AI's Good and Bad (1)
- Module 9: Risk, Resilience and Frontier AI (1)
- Module 2: Algorithm Filing and Registration (1)
- Module 3: High-Risk AI Requirements: The Technical File (1)
Questions
- What does model validation cover in AI governance?
- 2 competencies are filed under it: Model risk management and independent challenge, AI evaluation and testing design. Each has its own page saying what it means in practice.
- Which roles ask for model validation?
- 15 roles on the map name at least one of its competencies, from analyst seats to executive ones. Each role page lists the depth expected.
- Where is model validation taught and graded?
- 32 graded topics teach it across 9 programs. Each topic is graded by explaining it back against its own transcript, so a pass is evidence, not attendance. The first module of every program is free with a free account.