The most common mistake I see boards make with AI isn't underestimating the technology. It's misplacing it. AI gets treated as another IT project — something budgeted for, deployed by the IT department, and reported back as an implementation status update. That's a natural reaction, because AI enters the organisation through its technical layer. But it misjudges what AI actually is.
AI doesn't just change which systems you use. It changes how work gets done, which decisions can be automated, which skills become valuable, and which risks the organisation takes on — legally, ethically and reputationally. It isn't an infrastructure decision. It's a strategic decision that happens to also require infrastructure.
Five questions I believe every board should be able to answer
- Where in our business does AI create real value — not potential, but real — and where are we simply along for the ride because everyone else is?
- Who is accountable when an AI model makes a decision that turns out to be wrong? Is that clear to the organisation, or do we only find out when it happens?
- What happens to our data governance when more systems are trained on, and act on, data we previously treated as static?
- What do we do about the part of the organisation whose tasks become redundant or significantly changed — do we have a plan for transition, or do we find out too late?
- Are we investing enough — or too much, too fast, without a clear hypothesis for where the value is actually meant to come from?
None of these questions are answered by a technical project plan. They are answered by a board that understands the technology deeply enough to ask the CIO and CTO the right questions — while understanding the business well enough to translate the answers into strategy, risk and capital allocation.
Why it takes both at once
I've spent most of my career at the intersection of technology and business — from building an international security company from the ground up to now serving on boards that need to navigate this same type of decision. That experience has taught me that the limiting factor is rarely the technology itself. It's the ability to translate it into something a board can actually act on.
It isn't the board's job to be a technical expert in AI. It's the board's job to make sure the organisation asks itself the right questions — before competitors or the market ask them for you.
— David Hald