Glossary term
AI Governance
Subject: Technology, Cyber and Data, AI Governance and Risk
Also known as: Artificial Intelligence Governance
AI governance is the framework of policies, roles, approvals and oversight that decides how an organisation builds, buys and uses AI, and who answers for the results.
Explanation
AI governance gives an organisation a clear answer to who may deploy AI, for what purpose, under whose approval and with what checks. Typical parts are a policy that sets principles and banned uses, a named owner for each system, a risk-based approval step before launch, standards for data and testing, monitoring after launch, and a route for escalating incidents to senior management or the board. It usually sits next to existing data, model risk and technology governance instead of replacing them. Laws such as the EU AI Act and standards such as ISO/IEC 42001 now shape what the framework must cover. A committee with members from legal, risk, data science and the business often runs it. A regulator's first question after an AI failure is who approved the system, and this framework is what supplies the answer.
Example
A multinational finds that three departments use different AI transcription tools, none reviewed by the IT or privacy teams. The board asks for an AI governance framework. Within a quarter the company names an AI committee, lists every tool in a register, bans uploading client contracts to unapproved services, and requires a risk review before any new tool goes live.
Not to be confused with
- Data Governance
- Data governance manages data quality, ownership and use; AI governance covers the systems built on that data, including their decisions and behaviour.