Perspective · AI Governance

Practical AI governance for mid-size organisations

You do not need a large ethics office to govern AI responsibly. You need a register, owners and a review rhythm.

In short: A mid-size organisation can govern AI responsibly with three habits: keep a register of every AI system in use, assign a named owner accountable for each, and review incidents, performance and data use every quarter. Add a short approval step for new high-risk uses.

Large companies set up AI ethics boards and dedicated teams. Most mid-size organisations cannot, and do not need to. What they need is a light, regular discipline that keeps track of where AI is used and who answers for it.

Keep a register of every AI system in use

List each AI tool or feature in use, including those built into existing software. For each, record what it does, which data it uses, who uses it, and the vendor. Many organisations find more AI in use than they expected once they look.

Assign an accountable owner to each

Every entry needs a named business owner who understands what the system is for, reviews its results and can switch it off. Ownership should sit with the function using the system, supported by IT and legal, not with IT alone.

Classify risk simply

  • Low: internal productivity tools that do not use personal or confidential data.
  • Medium: tools that use confidential data or draft external communications, with human review.
  • High: anything that affects decisions about people, money or safety. These need approval before use and closer monitoring.

Review incidents and drift every quarter

Once a quarter, owners report errors, complaints, changes in accuracy and any change in the data used. The review decides whether to continue, adjust or stop each system. Record the decisions; they are your evidence if a regulator or customer asks.

Keep up with the rules

Data-protection law, including India's Digital Personal Data Protection Act, already applies to AI that processes personal data. Sector regulators are adding guidance. A register and review rhythm make it much easier to respond as rules change.

Frequently asked questions

What is AI governance?

It is the set of policies, roles and reviews an organisation uses to make sure its AI systems are used safely, lawfully and for their intended purpose.

What should an AI register include?

The system's purpose, data used, users, vendor, risk level, accountable owner and the date of its last review.

Does the DPDP Act apply to AI?

Yes, where AI systems process digital personal data. Organisations must have a lawful basis, protect the data and respect individuals' rights.

About the author

Digital & Consulting Practice

Nuclear Edge

The team that plans and builds digital channels, business systems, cloud and AI workflows for growing businesses and enterprises.