Module 06 Lesson 2 · 7 lessons in this module

Governance Principles

In brief: Governance principles are not always perfectly compatible in every situation, and organizations benefit from deciding in advance how conflicts between them will be resolved. A useful approach is to treat accountability and transparency as non-negotiable baselines that apply everywhere, while letting proportionality determine the intensity of consistency…

Watch: How governance principles guide decision-making when conflicts arise, how proportionality enables AI risk tiering, and why every policy, role, and operating model in this module should trace back to the five core principles.

Module support notes

Using Principles to Resolve Governance Disputes

Governance principles are not always perfectly compatible in every situation, and organizations benefit from deciding in advance how conflicts between them will be resolved. A useful approach is to treat accountability and transparency as non-negotiable baselines that apply everywhere, while letting proportionality determine the intensity of consistency and documentation requirements for any given domain.

This avoids ad hoc debates every time a new governance question arises — and it gives stewards and domain owners a principled basis for resolving disputes without escalating every edge case to senior leadership.

Tip

When principles appear to conflict, use proportionality as the tiebreaker for scope and intensity — but never let it override accountability and transparency entirely. A low-risk domain may require lighter documentation depth, but it still requires documented ownership. The principle hierarchy should be explicit in the governance charter so every stakeholder knows how to resolve conflicts without needing a ruling.

A scenario where transparency and proportionality appear to conflict over documentation requirements for a low-risk domain. Resolution shown: proportionality governs documentation depth while transparency governs documentation existence — both principles are honored simultaneously by separating their scope.

Principles sometimes pull in different directions — organizations should decide in advance how they will be prioritized.

Principles as the Basis for AI Risk Tiering

The proportionality principle is most visible in how organizations tier their AI use cases by risk and apply governance rigor accordingly. This tiering approach — increasingly common in organizations subject to emerging AI regulation — allows governance teams to focus their most intensive review processes on high-consequence models while keeping low-risk analytical work moving without unnecessary friction.

Tier 1 — Low Risk: internal reporting and analytics models; light governance, no pre-deployment certification required
Tier 2 — Medium Risk: customer-facing recommendation models; moderate governance with quarterly certification reviews
Tier 3 — High Risk: credit, hiring, or safety-related models; heavy governance with mandatory pre-deployment certification and continuous monitoring

Warning

Applying the same governance intensity to every AI use case regardless of risk creates two problems: it buries high-risk models in the same review queue as low-stakes analytics, and it creates so much governance overhead for routine work that teams begin routing around the process. Proportionate governance is not less rigorous — it is more intelligent about where rigor is applied.

A three-tier AI risk classification showing Tier 1 — Low Risk for internal analytics models with light governance; Tier 2 — Medium Risk for customer-facing recommendation models with moderate governance and quarterly certification; and Tier 3 — High Risk for credit, hiring, or safety-related models with heavy governance, mandatory pre-deployment certification, and continuous monitoring.

Proportionate governance lets organizations focus the most rigorous controls on the AI use cases with the most consequence.

Every policy, role, and operating model in this module should trace back to these five principles — they are the foundation that keeps governance coherent as the program scales.

A circular diagram with the five governance principles, each labeled with its core definition and its specific AI governance application, showing how the principles work together as a coherent framework rather than as independent rules.

Every policy, role, and operating model in this module should trace back to these five principles.

Lesson progress

0% watched

← Previous Next lesson →