Executive Leadership and the Board | Blane Canada

Question 1

How much does a CEO actually need to understand about AI?

Accepted Answer

Enough to lead the people who do the technical work. You do not need to be a prompt engineer or a data scientist. You do need enough fluency to tell the difference between a smart investment and an expensive experiment, and to know when your team is solving a business problem versus chasing a technology fad.

You were probably selected to lead because you mastered the fundamentals of business and your industry: finance, marketing, operations, people, and industry trends. Those are still essential. They are just no longer sufficient, because every one of those disciplines is being reshaped by AI. Marketing is not just marketing anymore. Operations is not just operations. HR is not just HR.

Today a chief executive has to speak three languages fluently. Business. Technology. And the intersection between the two. That third language is quickly becoming the most valuable, and it is also the one very few executives have had any opportunity to learn.

The bar is fluency, not technical skill. You need enough understanding to know when your technical teams are making sound choices and when they are chasing novelty. You need enough to recognize where AI belongs in your business and where it does not.

Success here begins with literacy, not delegation. An executive who hands AI to someone else entirely loses the ability to evaluate what they are later told.

Question 2

What does my board expect me to know about AI?

Accepted Answer

Not expertise. Judgment. Your board does not expect you to understand large language models or AI architecture. They expect you to know what this means for the business, where the risk is, whether you have a direction, and whether you can answer when asked. That is an executive problem, not a technology problem.

This distinction is where most executives get into trouble. They measure themselves against a standard they were never asked to meet, and miss the one they are actually being evaluated on.

The board is not asking whether you can explain how a model works. They are asking something harder. Do you understand what this means for the business? Do you know where the risk is? Do you have a direction? Can you answer when they ask?

The leaders navigating this well are not the ones who mastered every AI tool. They are the ones who got serious about their leadership infrastructure: the right questions to ask, the right frameworks for evaluating decisions, and the right advisors to bring clarity where blind spots exist.

Your board does not expect you to know everything. It does expect you to know where AI belongs, and to make good decisions about it. That expectation is only going to get heavier over the next few years.

The board is watching. Not for expertise. For confidence, direction, and judgment.

Question 3

Who should own AI in our organization?

Accepted Answer

Someone with real authority who understands both the business and the technology landscape. In four decades of watching technology shifts, four versions of this decision play out and only one works. The other three fail predictably.

The most consequential AI decision most executives make has nothing to do with technology. It is about who they hand the responsibility to.

Version one: delegation to someone with real authority and a genuine grasp of both the business and the technology landscape. This works. It is also uncommon.

Version two: delegation to a trusted executive who has neither the background nor the experience to navigate these decisions. The organization moves, checks the box, and six months later nothing has changed, because the person in charge cannot tell the difference between a real opportunity and a vendor pitch.

Version three: no delegation at all, dressed up as strategy. Each department handles AI in its own area. Procurement decides for procurement, marketing for marketing. The result is a fragmented collection of disconnected tools, no coherent direction, and an executive team that genuinely believes it is moving forward.

Version four: handing the assignment to IT. Technology teams are essential to implementation. They are not equipped to lead strategy. Confusing those two responsibilities is one of the most consistent and costly mistakes I see.

The question is not whether your organization is doing something with AI. Most are. The question is whether the person responsible has the authority to make decisions that matter and the background to know which decisions those are.

Question 4

How do I answer the board when they ask what we are doing about AI?

Accepted Answer

Answer with a foundation, not an announcement. A tool launch has no defense when it underperforms. A diagnostic that identifies where value sits, a policy that bounds the risk, and a training plan that builds capacity gives you a defensible position and a measurable next step.

The question arrives in almost every boardroom now, and improvisation as a management strategy will only carry a leadership team so far when it does.

The instinct is to move fast. Pick a tool, launch a pilot, show progress. But speed without diagnosis produces the same failures that have stalled technology projects for decades, just on a compressed timeline. In my experience, what used to take eighteen months to go wrong now takes closer to three.

A better answer sounds like this. Here is where we examined our operation and what we found. Here are the two or three places where AI or a process change produces measurable value, ranked by impact. Here is what our governance exposure looks like today and what we are doing about it. Here is where our team's capability sits and what we are investing to close the gap. Here is what we expect to have completed and measured by the next meeting.

That answer holds up under follow-up questions. A chatbot announcement does not.

Question 5

Should I just delegate AI to my team?

Accepted Answer

Delegate the work, not the judgment. The executives struggling most with AI are not the ones who know the least. They are the ones who delegated it completely and then lost the ability to evaluate what they were being told. Uninformed delegation is the failure mode, not delegation itself.

I have seen this before. Not with AI. With ERP systems in the nineties. With CRM platforms in the two thousands. With cloud migration in the twenty-tens. The pattern is identical.

A leader decides this is technical work and hands it to the team. The team makes implementation decisions without adequate strategic context. Six months later the leader sits in a review meeting unable to tell whether the report they are hearing is accurate.

There is a specific kind of strategic literacy executives need here. Not technical fluency. Not the ability to write a prompt or understand a model architecture. The ability to ask the questions that keep the work honest.

Stay in the room. The specific questions worth asking are covered separately, but the principle is simple: delegate execution, keep evaluation. The moment you can no longer tell whether a report you are hearing is accurate, you have delegated something you cannot get back without starting over.

Question 6

Should IT own our AI strategy?

Accepted Answer

IT should own AI infrastructure, not AI strategy. Those are different responsibilities. Governance of AI decisions requires cross-functional authority and strategic judgment about the business, which is not what a technology team is structured or resourced to provide.

You cannot build an AI-driven business if AI is stuck in the IT department. AI needs to reach HR, finance, the shop floor, and the boardroom. If people touch it, they understand it.

The distinction that matters is between infrastructure and direction. Deciding which model to run, how to secure it, and how it connects to your systems is technology work and IT should own it. Deciding which business problem is worth solving, what the return has to be to justify the spend, and what an acceptable risk looks like is executive work. Handing the second set of decisions to the team that owns the first set is how organizations end up with well-built systems nobody needed.

There is a related question about who owns governance across the company, which is a separate problem with its own answer.

Question 7

If AI can answer almost any question, what is left for an executive to do?

Accepted Answer

Ask better questions. In a landscape where AI can produce a plausible answer to almost anything, the scarce and valuable thing is no longer the answer. It is the question, asked by someone who understands the business deeply enough to know what a good answer actually looks like.

Most executives have spent their careers being the person with the answers. AI is changing that job description, not because answers matter less, but because they have become cheap.

I built my own career on that distinction. For 22 years I ran a platform that helped organizations make better decisions about business retention and economic development. The system worked not because it had better data, but because it asked better questions than anyone had asked before it.

The executives leading well right now are not the ones who handed AI to a task force and waited for a report. They stayed in the room and got skilled at asking what a machine cannot ask for them. What problem are we actually solving? What does this outcome mean for our people? What are we assuming that we should be testing? Where is the risk that is not showing up in this recommendation?

What makes the question valuable is not cleverness. It is that you know things the machine has no access to. The history of why the last attempt failed. Which customer relationship is fragile right now. Which of your people is carrying more than the org chart suggests. What the competitive environment actually rewards. A machine cannot know what it was never told, and most of what matters in a business was never written down anywhere it could read.

Your job is not to know more than the machine. Your job is to know more about the business than the machine ever will.

The best title for the next chapter of executive leadership may not be CEO. It may be Chief Question Officer.

Question 8

What questions should I be asking my team about our AI projects?

Accepted Answer

Four questions keep any AI initiative honest. What problem is this actually solving? What would we lose if it failed? Who decided this was the priority, and on what basis? Where is the risk that is not showing up in this recommendation?

Those four do not require technical knowledge. They require a leader who stayed close enough to the work to know what to ask.

The first question is the one most often skipped, and it is the one that catches the most expensive mistakes. A team can describe what a tool does in great detail and still not be able to state the business problem it solves in one sentence. When that happens, you have found a solution looking for a place to land.

The second question forces an honest look at dependency and downside while it is still cheap to have that conversation. If this system failed next quarter, what stops working, who notices first, and how long would it take to go back to how you operate today? Teams rarely think this through voluntarily, because the answer is uncomfortable and the project is exciting.

The third question surfaces whether the priority came from a business case or from whoever presented most recently. Both happen in every organization. Only one of them is defensible when someone asks why this project and not the other three, and you can usually tell which you are dealing with by how quickly the answer arrives.

The fourth question is the one that protects reputations, including yours. Every recommendation presents its upside in detail. Very few volunteer the risk that did not make it onto the slide, not because anyone is hiding it, but because the people building something are the least likely to see what it exposes.

An executive who can ask a sharp question at the right moment is worth more to an AI initiative than an executive who built the system but cannot connect it to a business outcome.

Question 9

Am I the right leader to take our company through this?

Accepted Answer

It is a fair question and a rare one to ask honestly. The issue is not capability or commitment, both of which can be developed. It is fit: cognitive agility, learning velocity, and tolerance for ambiguity at the pace this moment requires. Those are diagnosable and they do not show up in a performance review.

AI is not just reshaping organizations. It is forcing an uncomfortable question in the C-suite.

Look at the pattern in 2026. An unusual concentration of major CEO transitions, with the reins largely going to chief operating officers who already knew the company and could accelerate without tearing down what was built. Doug McMillon cited urgency around AI as a major catalyst and said he did not think he could finish the transformation in time. James Quincey said essentially the same at Coca-Cola. One leadership consultant framed it well: every company has to run a faster race each year, and you need someone fit for that race who also has the longevity to see it through.

These are not turnarounds. They are transformations, and the destination is clear. The speed required to reach it simply outpaced the leader in the seat.

The leaders stepping aside are not abandoning their legacies. They are protecting them. Staying past your point of fit does not preserve what you built. It erodes it.

This is also something a structured outside assessment can surface, and it is often the most valuable thing it produces. Not a performance review and not a capability audit. An honest look at whether the leadership structure is fit for the race actually being run, at the speed it actually requires. The question sits underneath all four assessment dimensions: are the right people in the right roles to move as fast as this moment demands?

The harder question is whether your organization is asking the same thing before it becomes obvious from the outside.

Question 10

My team is ready to move on AI but my boss will not engage. What do I do?

Accepted Answer

You have hit the choke point, and it is a structure problem rather than a leadership failure. Executives need a way into this conversation that is not a pitch. A diagnostic that maps opportunity, cost, and risk changes the conversation from "I have an idea" to a discussion about organizational strategy.

I see this pattern constantly, and it comes in three versions.

The mood-dependent ask. "Is today a good day to bring up the AI build I have been working on?" The project is ready, the case is solid, and everything depends on timing.

The perception problem. Staff believe the boss is skeptical, so the conversation never starts. Sometimes the boss is skeptical. Sometimes nobody has ever given them a reason not to be.

The outright opposition. "Our CEO just does not like AI." No specific reason. No incident. Just a posture that closes the door before anyone knocks.

Every one of these is understandable. No executive wants to be embarrassed. No executive wants to experiment without some assurance the investment makes sense. And when the board is watching, "we tried an AI thing" is not a reassuring update.

The choke point is not a leadership failure. It is a structure failure. What breaks it is a diagnostic that maps the opportunities, surfaces the costs, and identifies the risks before anyone commits to anything. With that in hand, you are no longer asking your boss to take a personal bet on your idea. You are handing them a decision they can defend.

Question 11

Do I need to be technical to lead an AI initiative?

Accepted Answer

No. You need to be a translator. The gap that kills these projects is not technical skill. It is the space between what AI can do and what your business actually needs. Someone has to stand in that gap, and it does not have to be a coder.

I have said "I am not a coder" hundreds of times across a 30-year career. My strengths were in product design, process organization, and problem solving for clients, then translating all of that into something developers could implement.

In football, games are won and lost in the gaps, in the space between the lines where plays break through. In business, there is a critical language gap between technology and the operation. Technical people and executives speak different languages, and that gap has compounded the difficulty of implementing AI everywhere I look. Executives are drowning in advice about tools, platforms, and dashboards, each promising change and most delivering confusion.

The problem is not the technology. The problem is translation.

You do not need to become a coder to get value from AI. You need someone who closes the gap between the technology and the business outcome. In a period like this one, the advantage belongs to the translators.

Question 12

Why can I not seem to ask the right questions about AI?

Accepted Answer

Because good questions come from a working model of what the thing actually does, and most executives have never been given one. Without it, you are evaluating recommendations on delivery rather than substance. Strategy then rests on assumptions nobody tested, and governance never gets written because nobody knows what they are governing.

This is the honest version of a problem a lot of executives will not say out loud.

If you cannot see the big picture, you do not know what questions to ask. And without the right questions, everything downstream degrades. Strategy is built on assumptions nobody tested. Execution chases whatever was demonstrated most recently. Governance never gets written because nobody knows what they are governing.

The fix is not becoming technical. It is acquiring enough of a working model to evaluate what you are shown. Once you understand roughly how these systems arrive at an answer, the questions change on their own. You stop asking whether the demonstration was impressive and start asking what it was given, what it was not given, and what it would do with the messier version of the same input.

There is also a practical test you can run on any AI recommendation without knowing anything technical. What specifically improves if this succeeds: revenue, margin, capacity, response time, decision quality, or risk? Is the information it needs actually available? Were the people affected involved? What does full implementation cost, rather than the prototype? What alternatives were considered? Who owns the outcome, and what would cause us to stop? Working through those is usually worth more than another demonstration.

That foundation is what turns an executive from a spectator into a buyer who can tell the difference between capability and performance.

Question 13

Why do top-down AI mandates struggle?

Accepted Answer

Because leadership sets direction without the operating detail. The people doing the work know where information actually comes from, which exceptions matter, and which undocumented workarounds keep the process running. Leave them out of discovery and those details never reach the requirements, so the system solves a process that does not exist.

Executive sponsorship is necessary. It is just not sufficient, and the gap between those two things is where a lot of money disappears.

We learned this on an engagement of our own. The scope was to turn a client's daily job schedule into optimized driver routes, which was a clean problem with a clean solution and completely wrong. That operation does not schedule day by day at all. Drivers receive a batch of work and route themselves through the week. The detail had surfaced in discovery, but not as a hard rule, and we only understood what it meant once we were building against it.

Delivering the original design would have produced a process the client did not need and would not have used. Technically complete, solving a problem that did not exist.

The strongest pattern combines both directions. Leadership defines why the problem matters and what outcome justifies the investment. The people doing the work explain how it actually happens, including the parts nobody documented. Then the builders translate. Skip the middle step and you get a system delivered to employees rather than built around their reality, which damages the design and the adoption at the same time.