Boards are being asked to oversee something that moves faster than their reporting cycle. Most directors are not AI specialists, and they do not need to be. The board's job is not to have the technical answers. It is to ask the questions that reveal whether management's strategy is sound, where the risk actually sits and whether the organisation is positioned to gain from the shift rather than just survive it.
The questions below do not require technical fluency. They require the same discipline boards already apply to capital allocation, audit and succession. What matters is less the polish of the first answers than whether management can support them with evidence.
One. Where does AI put the most pressure on our business model, and where does it open the most room? A sound answer names specific functions, revenue lines and customer segments. A weak answer talks about efficiency in general terms and cannot say which part of the P&L moves first.
Two. How does our exposure compare with our peers? If management cannot describe what comparable organisations are doing, that is not a secret competitors have kept. It is a research gap, and it is closable.
Three. What are we investing in, and why those things first? Sequencing is the strategy. An answer that lists initiatives without an order, or an order without a reason, is a portfolio, not a plan. A sound strategy can also say what it is deliberately not doing yet.
Four. What would have to be true for this strategy to fail? Every strategy rests on assumptions about technology, customers and competitors. If management cannot name theirs, the assumptions are still there, just unexamined.
Five. Who is accountable for AI in this organisation? A named executive with authority and budget is one answer. A working group that reports occasionally is a different answer, and the difference shows up the first time a hard trade-off arrives.
Six. How are we governing the AI already in use? In most organisations, staff adopted AI tools before any policy existed. The question is not whether that happened. It is whether management knows where organisational data is going and which decisions already have AI in the loop.
Seven. Which regulatory obligations apply to us now, and what is coming? The answer is jurisdiction and sector specific. What the board is listening for is whether management is tracking the obligations that apply to its own operations, or waiting for regulation to arrive as a surprise.
Eight. How do we know the AI outputs we rely on are reliable? Anything that reaches customers, regulators or financial reporting needs a review step someone owns. The useful follow-up is simple: who checked, and how would we know if it was wrong?
Nine. What is our workforce plan, on both sides of the ledger? AI changes roles and it creates capability needs. A plan that only counts efficiency, or only promises reskilling, is half a plan. The organisations that come out ahead treat the capability build as seriously as the cost line.
Ten. What early signals are we watching, and who reads them? Competitor moves, hiring shifts, new entrants and regulatory signals all arrive before the annual strategy refresh does. Someone should own the watching, and the board should hear about it more than once a year.
None of these questions requires the board to run the strategy. They establish whether one exists, whether it rests on evidence and whether the organisation is set up to act on it. A shared, researched baseline of where the organisation is exposed and where it holds advantage makes every one of these conversations shorter and sharper. That baseline is exactly the kind of thing a board can reasonably ask management to put on the table.
What would your organisation’s score tell the board?
A Snapshot gives the board a shared, researched baseline.
Analyse my organisation →