About GovCompass

AI governance, made usable.

GovCompass turns the frameworks that govern AI, the EU AI Act, the NIST AI Risk Management Framework, ISO/IEC 42001, the OECD principles and the GDPR, into the decisions real teams have to make. It does that through two layers: a free knowledge base and a paid, structured training.

AI governance, free

Plain-language answers on the EU AI Act, the GDPR and the standards that govern AI, organized around the seven pillars of responsible AI. Guides, analyses and legal references, open to everyone.

Explore AI governance

Academy, paid

AI Governance Academy: deep, original lessons, scenario-based questions and honest readiness tracking. The first track prepares you for the AIGP exam; more frameworks follow.

See the Academy

The seven pillars

GovCompass uses the seven pillars of responsible AI to organize the knowledge base and the Academy. The structure helps connect related requirements across sources. It does not replace them.

GovCompass is an independent knowledge base. It does not issue a standard, certify conformity, or claim formal authority for the way it organizes its content. What it adds is the internal control and assurance view: how a requirement becomes a control, how that control is tested, and what evidence shows that it operates.

Fairness
Safety & reliability
Privacy
Security & robustness
Transparency & explainability
Accountability
Human oversight

Who writes this

Michel Venniker

Lead AI Transaction Monitoring

Michel Venniker is Lead AI Transaction Monitoring within Financial Crime Compliance Controls at a major Dutch bank, where he advises on and challenges the development and deployment of machine learning models for transaction monitoring, with a focus on transparency, explainability, quality assurance, and governance for these models. He has over 25 years of experience in IT audit, financial crime compliance, and risk and control at major banks. He is the founder of GovCompass, an independent knowledge base on AI governance.

How the content is made and checked

Original wording throughout: nothing copied or paraphrased from a textbook.
Grounded in primary sources: the EU AI Act, OECD principles, NIST AI RMF and the official exam blueprint.
Written to one register: plain, literal English at exam level, one idea per step.
Every lesson passes a quality checklist before it is published: coverage, a worked example, and feedback that teaches.

Explore AI governance See the Academy