A global view of AI governance

Governing AI is a human choice.

AI systems are reshaping decisions, institutions and public life. This concise guide maps the questions that matter: who decides, who benefits, who is accountable and how rights are protected.

Explore the field

01 / Overview

AI governance,
in plain language.

AI governance is the set of rules, institutions, practices and technical controls used to shape how AI is designed, deployed and overseen. It connects public values with decisions made in real systems.

R

Rights

Protect people from discrimination, manipulation, unsafe systems and unjustified surveillance.

R

Responsibility

Make roles, decisions and remedies clear across developers, deployers and public authorities.

R

Resilience

Test systems, monitor impacts and respond when technologies, uses or risks change.

02 / The field

From principle
to practice.

AI governance works across three connected layers.

  1. 01

    Law and policy

    Legislation, standards and public institutions define duties, limits and paths to redress.

  2. 02

    Organizations

    Leadership, procurement, risk ownership and human oversight turn principles into operating decisions.

  3. 03

    Systems

    Documentation, evaluation, monitoring and incident response provide evidence throughout the AI lifecycle.

03 / Key questions

Five questions every AI system should answer.

  1. 01

    Purpose — What is this system for, and should AI be used here?

  2. 02

    Impact — Who can be helped, harmed or excluded?

  3. 03

    Evidence — What shows the system works in its real context?

  4. 04

    Accountability — Who can stop it, challenge it and repair harm?

  5. 05

    Change — How will risks be monitored as the system, data and environment evolve?

A practical starting point

Good decisions begin with better questions.

Effective governance is not a one-time checklist. It connects law, organizational judgment and technical evidence—and it must evolve as systems and their effects change.