Some claim that AI is the biggest disruptor we’ve ever seen. Which is why so many business leaders are under immense pressure to navigate this new world order and, critically, determine how best to immerse the business in it while remaining aware of all the risks. And there are many: misaligned expectations, poor data foundations, unclear governance, ethical concerns, and solutions that never make it beyond the pilot stage.

It’s akin to finding the best path through a proverbial minefield but blindfolded with your hands tied behind your back!

That might seem a bit extreme and plenty of organisations have successfully integrated AI. However, from my conversations with peers and even from conferences I’ve attended, there is still a lot of nervousness out there; still a lot of people are unsure and concerned about making any choice that might come back to haunt them.

With that in mind, I’ve put together a checklist for business leaders responsible for managing an AI initiative who want clarity and a sense of direction amid all the noise. It’s a checklist I’ve developed from working with several of our clients and from internal work as we’ve integrated AI into our workflows and business processes.

AI is not a strategy; it's an enabler. If your idea is to hop on the AI train without a destination or even a purpose for your journey, then the results will reflect this lack of direction. 

The organisations seeing the greatest returns are those that begin with a precise business problem, not a technology mandate.

Therefore, before committing resources, your leadership team must establish the following targets:

• What specific operational or customer outcomes are we targeting?
• How does this AI initiative support our 3–5 year transformation roadmap?
• What does success mean for us, and how will we measure it?

If your answers are vague, your AI programme will be too. Strategic clarity is the foundation on which everything else is built.

From experience, I think this is the hurdle that many organisations struggle to overcome. Research validates this, with 87% of CIOs saying data challenges hold back AI ambitions.  And for the larger ones, or those in the public sphere with multiple departments and siloed teams, centralising data can be a real challenge.  If this is you, my advice is to take a breath and then prioritise data management. Remember, what you get out is only as good as what you put in, so if it’s not in optimum shape, what you get will set you up for failure.

Assess your position:

• Is your data centralised, accessible, and well-governed oris it siloed across systems and business units?
• Do you have clear data ownership and lineage across your most critical datasets?
• Are you compliant with data privacy regulations in every market you operate in?

For each of these, you need to invest the time to properly answer, then clarify that all your data is ready and in the best possible shape. 

While AI and the technology behind it are not new, the speed at which it is overtaking business operations is, and so organisations are scrambling to keep up. And people are too. In every company, there will be the outliers, the employees who are at the head of the pack. But there aren't usually enough of them, so there will typically be vast swathes of people who are just dipping their toes into the new tech, and so are very limited in their knowledge and expertise.

As you consider your people, ask yourself:

• Do we have AI/ML expertise in-house, or will we partner with or outsource to an agency?
• Are our business leaders sufficiently AI-literate to make sound governance decisions? (Yes, this can be a tough one to ask, let alone answer, as people don’t necessarily want to reveal their lack of familiarity with something as important as AI. So this has to go beyond egos or nervousness and get to the truth.)
• Have we given any consideration to the employees most affected? Will we offer L&D and upskilling opportunities?

I’m very aware that this topic can be rather sensitive, and people are rightly worried about their jobs. It needs to be approached with a high level of EQ and transparency. Meet with your teams, give your employees access to L&D resourcing, and, when appropriate, look elsewhere to agencies like ours that have the talent and skills.

As we discussed in AI Governance in 2025: What you need to know, regulatory pressure on AI is intensifying. And although there is some disagreement about how far to go, the thing to remember is that, regardless of the law, your customers want to see responsible AI practices. They want transparency. And your employees do too. It’s in everyone’s interest for you to have defined protocols and to have a clear understanding of your legal obligations.

Ask yourself:

• Do we have a defined AI ethics policy and a cross-functional team to enforce it?
• Are there clear accountability structures for AI-driven decisions, particularly high-stakes ones?
• How are we monitoring deployed AI systems for bias, drift, or unintended consequences?

Governance is essential, not just a routine step. It provides the foundation to innovate at scale while protecting your reputation and reducing legal and operational risks.

Our CTO would probably suggest I put this first because, all too often, in discussions with business leaders, it becomes clear that many legacy technology stacks simply cannot support the compute, integration, or scalability that enterprise AI demands.

So ask your CTO or IT team:

• Is our cloud and compute infrastructure sufficient for the workloads we intend to run?
• Can our existing systems integrate with AI platforms via modern APIs without prohibitive cost or complexity?
• Do we have the security architecture in place to protect AI models, training data, and outputs?

Along with the answers to these questions, you need the time and resources to address any issues that arise.

Be deliberate. Yes, business leaders are under a lot of pressure to deploy AI, and it might be tempting to roll something out to get things moving.

However, to do so would be a mistake. Use this checklist not as a box-ticking exercise, but as a diagnostic. Where are your gaps? Where are you most exposed? Sort these out first, and take the time to ensure you have all the measures in place so that when you deploy AI, it will deliver the transformation you are promising your board, your customers, and your people.