Non-profit

AI governance for non-profit organisations

A non-profit board does not need to understand machine learning. It needs to understand who is accountable when an AI system makes a decision that affects donors, beneficiaries, or staff.

By Zakir Hoosen, Director, Fuzzelogic Solutions. Board-level guidance in plain English.

Governance is not a policy document sitting in a shared drive. It is a decision about who answers when the system gets it wrong. For a non-profit, that question matters more than most. You are handling donor money, beneficiary data, and public trust. A mistake does not just cost you money. It costs you credibility.

Most non-profit boards I sit with do not have an AI problem. They have a governance gap. AI is already inside the organisation, often introduced by a well-meaning team member who signed up for a free tool and started feeding it data. The board has not seen it, has not approved it, and does not know it is there.

This article is about closing that gap.

What governance actually means

AI governance is not complicated. It is a set of decisions about what the organisation will and will not let AI do, who approves each use, and what happens when something goes wrong. If those decisions have not been made, you do not have governance. You have hope.

1. A governance framework should answer four questions:
2. What data goes in?
3. What decision comes out?
4. Who is accountable for that decision?
5. What happens when it is wrong?

That is it. Everything else is detail. If a board cannot answer those four questions for every AI system in use, there is a problem.

The research is blunt

The gap between boards and AI governance is not a non-profit problem alone. It is an everywhere problem.

One in five organisations. No governance. And that number is from enterprises with resources most charities would envy. For non-profits, where teams are smaller and budgets are tighter, the gap is likely wider.

A board that rushes and lacks an informed view is a board that will wave through something it cannot challenge. That is where risk lives.

Why non-profits face a harder version

Non-profits carry a specific burden that commercial businesses do not. You hold data on vulnerable people. You distribute funds on behalf of donors. You answer to regulators, to the Charity Commission in the UK, and in the Isle of Man to the Charities Registration Office and the broader framework the government applies to non-profit governance.

The Isle of Man government has been clear that charities operating on the island are expected to meet the same standards of accountability and transparency as any other regulated entity. That expectation does not stop at technology. If a charity uses AI to triage applications, decide who receives funding, or communicate with beneficiaries, the governance around that system needs to be as robust as the governance around the finances.

The Joint Fiduciary Standards Commission in the Isle of Man issued guidance on AI governance in July 2026. It was directed at financial services, but the principles apply to any organisation handling sensitive data under Manx regulation. Know what the system does. Know what data it uses. Know who is responsible. That is not optional guidance. It is the direction of travel for every regulator on the island.

The four questions a board should ask

When AI comes up at a board meeting, four questions are enough.

First, who owns this? Not a team. Not a committee. A person. The person who answers when it goes wrong.

Second, what data does it use? Where does it come from? Who checked it last? If the answer is we are not sure, stop there.

Third, what decision does it make? Not what the tool does in general. What it does for your organisation, with your data, right now.

Fourth, what happens when it is wrong? What is the process? Who is notified? Who is accountable? If the answer is we do not know, you do not have governance. You have exposure.

Two thirds failed. Not because the technology was wrong. Because the governance around it was absent.

The practical steps

Governance does not require a large budget or a new hire. It requires a register. Every AI system in the organisation, including the ones no one told the board about, needs to be on a single list. For each one, record what it does, what data it touches, who approved it, and who is responsible.

That register becomes your governance framework in practice. It is the thing a new trustee can read in ten minutes. It is the thing an auditor can review. It is the thing that protects the organisation when a regulator asks questions.

Fuzzelogic uses a framework of five tests for AI readiness. Reachable, Trustworthy, Explainable, Changeable, Governed. Every system in that register should pass all five. If it does not, it is not ready for board approval.

The honest version

Fuzzelogic is an Isle of Man firm that has spent nineteen years modernising systems for boards across banking, insurance, healthcare, and government. We tell non-profit boards what most consultants will not. The honest answer is often that AI should not touch a particular process at all. When that is the case, we put it in writing rather than build it anyway.

Your systems were built for a world before AI. Most can get there. We tell you which ones cannot.

Start with the assessment. Two to four weeks, fixed price, and you own the verdict and the roadmap whether or not we build any of it. When you are ready to talk AI, call Fuzzelogic Solutions and ask for Zak. www.FuzzelogicSolutions.com | info@FuzzelogicSolutions.com | +44 (0)1624 618950