AI Basics

AI readiness scorecard: 10 questions every board should ask

Ten questions. Ten honest answers. The scorecard tells you where your business stands and what needs to change before AI can deliver.

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

Most boards think they are ready for AI. Most are not. The difference is not the budget or the ambition. It is the data, the governance, and the process change. This scorecard gives you ten questions. Answer them honestly and you will know where your business stands.

The scorecard is not a test you pass or fail. It is a map. It shows you where you are ready and where you are not. That is useful information. It is the information most boards do not have.

The ten questions

Score each question from one to five. One means you cannot answer it at all. Five means you have it documented, tested, and enforced.

1. Where does the data the AI needs sit, and can it be found when the system needs it?
2. Is the data accurate, current, and complete, and do you know that it is?
3. Can someone explain why the AI made a particular decision, in plain English?
4. Can the system be changed when the business, the market, or the regulation changes?
5. Has someone been appointed to govern each AI system, with the authority to stop it?
6. Do you know where AI already exists in your business, including the tools no one told the board about?
7. Has the board seen a cost-benefit analysis in numbers the finance director would accept?
8. Do you have a plan for how the process will change, not just the technology?
9. Has the workforce been mapped, and do you know which jobs will change?
10. Do you have a regulatory plan that covers the JFSC, the GFSC, and any cross-border requirements?

Score yourself. Be honest. The score is not for anyone else. It is for you.

What the scores mean

1. 40-50: You are ready. Start with a specific use case, govern it well, and move.
2. 30-39: You are close. Fix the gaps before you start. The gaps are governance, data, or process.
3. 20-29: You are not ready. You need preparation before any AI project. The preparation is weeks, not months.
4. 10-19: You have significant work to do. The data, the governance, and the process are not in place. Start there.

The scores are honest. Most businesses score between twenty and thirty-five. That means most businesses need preparation before they are ready. The preparation is not expensive. It is essential.

Why the questions matter

The questions are not random. Each one addresses a failure point that kills AI projects.

Question one, the data location, matters because AI cannot use data it cannot find. Most businesses have data in multiple systems. The AI needs it in one place. If the data is scattered, the AI will not work.

Question two, the data quality, matters because AI trained on bad data produces bad results. A business that does not know whether its data is accurate is a business that does not know whether its AI output is reliable.

Question three, the explanation, matters because governance requires it. The JFSC expects firms to explain AI decisions. If no one in the business can explain the output, the system is not compliant.

Question four, the changeability, matters because the business changes. The AI must change with it. A system that cannot be updated is a system that will be wrong soon.

Question five, the governance, matters because accountability requires a person, not a committee. If no one has been appointed, no one is accountable.

Question five is the one most businesses fail. The system exists. No one owns it. That is not governance. It is luck.

The Isle of Man context

Isle of Man businesses have specific constraints that affect readiness. The talent pool is small. The teams are lean. The regulatory framework is clear but demanding. These constraints make the questions more important, not less.

For Isle of Man firms with cross-border exposure, question ten is critical. The JFSC, the GFSC, and the EU AI Act all apply. The board must understand which regime governs which system and ensure compliance with all of them.

The Island's size is an advantage for readiness. A small business can answer these questions faster than a large one. Fewer systems. Fewer people. Fewer moving parts. The answers are simpler to find.

Readiness is part of the seventy percent. It is process change. It requires the business to understand itself before it adopts AI. That takes executive attention, not just a technology budget.

How to use the scorecard

First, score it honestly. Do not give yourself a four because you want a four. Give yourself the score that matches the reality. The score is for you, not for anyone else.

Second, identify the gaps. Where did you score lowest? That is where the work needs to happen first. The gaps are usually in governance, data, or process.

Third, make a plan. The plan should address the lowest scores first. Governance before technology. Data before algorithms. Process before tools.

Fourth, reassess. The scorecard is not a one-time exercise. The business changes. The scores change. Reassess quarterly.

The two thirds that failed often skipped the readiness check. They started with the technology. The scorecard tells you whether you are ready to start. Most businesses are not. The good news is that readiness takes weeks, not years.

The honest assessment

Here is what most consultants will not say. Some businesses should not score themselves. They should have someone independent do it. The business is too close to its own systems to see the gaps honestly.

If the honest answer is that you need an independent view, Fuzzelogic provides one. The assessment is two to four weeks, fixed price. You own the verdict and the roadmap whether or not we build any of it.

You already have AI in your business. You just do not know where. The scorecard tells you whether you are governing it. The honest answer is probably not.

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