Ali Fouad KwikaCTO · Dubai
All insights · 6 min read · Ali Fouad Kwika

AI Readiness Assessment: What It Covers and How to Run One

AI Readiness Assessment: What It Covers and How to Run One

An AI readiness assessment is a short, structured review that shows whether your company can put AI into real use, and what to fix first. It looks at six areas: business case, data, technology, people, governance and delivery. I am Ali Fouad Kwika, a Dubai-based CTO and AI consultant, and this is how I run one.

What an AI readiness assessment is, and what it is not

It is a decision tool. At the end you should know which AI use cases are worth funding now, which need groundwork first and which to drop. If the result does not change what you do next quarter, it was not an assessment.

The right time for one is before a large commitment: a platform licence, a new data team, or a pilot that the board expects to become a product.

The six areas to assess

1. Business case

Which decisions or tasks would AI improve, who owns each one, and how will you know it worked? A use case with no named owner and no measure is an experiment, not a project. Rank the candidates by value and by how hard they are to deliver, and start where both are in your favour.

2. Data

Where the data lives, who is allowed to use it, how complete and current it is, and whether it may lawfully be used for this purpose. Check a sample by hand: open the records the model would rely on and ask whether a new employee could make sense of them. If a person cannot, a model will not either.

3. Technology

Can your current systems send data to a model and receive an answer safely? Look at integration points, identity and access, logging, cost controls and where the workload would run. You rarely need new infrastructure to begin, but you do need to know what connects to what. My build vs buy framework for AI helps decide what to buy, what to assemble and what to build.

4. People and skills

Who will own the system after launch, who will check its output, and do the people whose work changes understand why? Readiness includes the staff who use the tool every day, not only the engineers. A company with nobody responsible for quality once the consultants leave is not ready, however good the model is.

5. Governance and risk

Which rules apply to the data and to the decision, who approves a new use case, how are errors caught, and what is the human fallback? In the UAE the answer depends on your sector and on whether you operate on the mainland or in a financial free zone. I cover this in which AI rules apply in the UAE, and companies that sell into Europe should also read whether the EU AI Act applies to UAE and GCC companies.

6. Delivery

Can you take something from pilot to production? Look for a way to test with real users, measure quality over time, release changes safely and switch the feature off if it misbehaves. A pilot that works in a demonstration can still stall here.

How to run the assessment

  1. Agree the question. Write down the decision the assessment must support, for example "should we fund these three use cases next year?"
  2. Talk to the people who do the work. Interview the owner of each process, the data and IT leads, and legal or compliance. Ask what they do today, not what they hope AI will do.
  3. Look at evidence. Sample the data, read the architecture diagrams, check access rights and existing contracts. Opinions in interviews and facts in systems often differ.
  4. Score each use case against the six areas. Red, amber or green per area is enough. The aim is to see what blocks each use case, not to produce a precise number.
  5. Decide. Sort the use cases into start now, prepare first and stop. For each "prepare first", name the gap, the owner and the next step.

What you should receive at the end

  • A ranked list of use cases, each with an owner and a success measure.
  • The gaps that block each one, in plain language.
  • A plan for the first 90 days: what to buy or build, what to fix in the data, and who is responsible.
  • A cost range and the assumptions behind it.
  • The risks and rules that apply, with the control for each.

Gaps that often appear

The same gaps tend to come up: the use case has enthusiasm but no owner; the data exists but sits in several systems with different definitions; nobody has decided who checks the output; and the pilot was never designed to be operated. None of these is a reason to stop. They are the work to do before the choice of model matters. My Doctor of Business Administration research at Clermont School of Business was on AI adoption in GCC firms, which is why I treat readiness as an organisational question as much as a technical one.

Do it in-house or bring someone in?

An internal team knows the business and the data best, and can run the assessment well if it has the time and someone senior enough to say what is not ready. An outside reviewer is useful when departments disagree, when the team is too close to a favourite project, or when the board or an investor wants an independent view. If you bring someone in, make sure they are not paid according to which tool you choose; how to choose an AI consultant in Dubai lists the questions to ask.

Frequently asked questions

What is an AI readiness assessment?

A structured review of whether a company can use AI usefully and safely. It covers business case, data, technology, people, governance and delivery, and ends in a decision about which use cases to start, prepare or stop.

Who should be involved?

The executive sponsor, the owners of the processes in question, the data and IT leads, legal or compliance, and a few of the staff who would use the tool every day.

Do we need clean data before we start with AI?

Not all of it. The data behind your first use case has to be good enough for that use case, and you need a plan for the rest. Waiting for perfect data across the company delays everything.

How is an AI readiness assessment different from an AI strategy?

The assessment shows where you stand and what is feasible now. A strategy sets where you want to be over several years. Do the assessment first, so the strategy starts from evidence.

Next step

If you want an independent AI readiness assessment, see how I work as an AI consultant in Dubai and the UAE, or how AI fits into wider digital transformation consulting. Consulting firms and agencies that need this for their own clients can work with me as a partner. To talk about your case, contact me.

Reviewed on 19 November 2026. This article describes a working method and states no time-sensitive figures; the linked articles on UAE and EU rules list their own sources and check dates.

Have a product to build or a team to scale?

Tell me where you are and where you want to be. Let’s find the shortest path to get there.