Most leaders talk about AI. Few actually study it. I decided to do both — and what I learned at Oxford changed how I think about leadership, technology, and the future of business forever.

In early 2023, I completed the Oxford Artificial Intelligence Programme through Oxford’s online learning platform, finishing with an overall grade of 88.9% — above the class average of 83.2%. But the grade is almost beside the point. What matters far more is what six intensive modules of deep, structured learning revealed about where AI is heading, how leaders should respond, and why most organisations are still asking the wrong questions.

This is not a course review. This is a leadership reflection.

The Conventional Wisdom

The prevailing view in most boardrooms goes something like this: “We need to adopt AI. We’ll hire a data science team, plug in some tools, and let the technology do the rest.” Leaders are told to be “AI-aware,” to attend a half-day workshop, to read a McKinsey report, and to delegate the hard thinking to someone with a PhD in machine learning.

The conventional wisdom says AI is a technical problem. It belongs in the IT department. Leadership’s job is to set the vision and get out of the way.

I used to half-believe this. Then I spent weeks inside the Oxford Artificial Intelligence Programme — and I no longer believe it at all.

A Different Perspective

AI is not a technical problem. It is a leadership problem. And leaders who refuse to truly understand it are not delegating responsibility — they are abdicating it.

Let me explain what studying this subject at Oxford’s level of rigour actually revealed.

The programme took me through machine learning, deep neural networks, recommendation systems, image recognition, ethical frameworks, and finally, building a full business case for AI implementation. Each module demanded not just comprehension but application — how does this work inside a real organisation, with real constraints, real people, and real consequences?

In Module 4, I submitted a project exploring AI’s potential within my own organisation. I proposed a sandboxed CI/CD pipeline approach for AI-assisted code changes — a concept my assessor described as “pretty complex and novel” with “applications in other industries.” That idea didn’t come from a textbook. It came from understanding the technology deeply enough to connect it to a specific operational challenge. You cannot make those connections from a distance. You cannot lead transformation you do not understand.

In Module 5, I developed a set of ethical principles for AI deployment, earning 100% — and more importantly, earning feedback that highlighted my emphasis on responsibility as a standout element. That is not an accident. Responsibility is a leadership word. It should sit at the centre of every AI strategy, not as a compliance checkbox, but as a cultural commitment.

“The most dangerous leader in the AI era is not the one who knows nothing about the technology. It is the one who knows just enough to be confidently wrong.”

Even my lower-scoring modules taught me something critical. In Module 1, I analysed why VHS beat Betamax — a case study about technology adoption that has direct parallels to AI today. My assessor pushed me to go deeper on accessibility and affordability. That feedback stung slightly, but it was correct. The best technology does not always win. The most accessible, most trusted, most human-friendly technology wins. That is an insight every AI leader needs tattooed on their wall.

In Module 3, I was pulled up for answering the question I wanted to answer rather than the one being asked — specifically around image recognition features. Again, a leadership lesson hiding inside a technical critique: listen precisely, respond precisely, and never assume you already know what the problem is.

Counterarguments — And Why They Fall Short

Some will argue that leaders don’t need to understand the mathematics of gradient descent or the specifics of q-learning versus modern reinforcement learning approaches. Fair point — and the Oxford programme itself would agree. The goal is not to turn executives into engineers.

But there is a significant gap between understanding everything and understanding enough. Right now, most senior leaders sit far too close to the “nothing” end of that spectrum. They cannot evaluate vendor claims. They cannot challenge their own data science teams. They cannot identify where AI will create genuine value versus where it will generate expensive noise. They cannot make ethical judgements about deployment without understanding what is actually being deployed.

Structured, rigorous education — not a podcast, not a LinkedIn post, not a conference keynote — is what closes that gap. The Oxford Artificial Intelligence Programme is built precisely to serve that purpose, and it delivers.

What This Means for Business Leaders

If you are in a leadership position in 2024 and beyond, here is what my Oxford experience tells me you need to face squarely:

• Your competitors are not waiting. AI is already reshaping recommendation engines, customer support, code development, and operational efficiency. The organisations pulling ahead are those where leadership understands the tools well enough to deploy them intelligently.
• Ethics is now a strategic function. The EU AI Act, growing public scrutiny, and high-profile AI failures mean that responsible AI governance is not optional. Leaders need the conceptual framework to design and enforce ethical principles — not just sign off on a policy document someone else wrote.
• Your instincts about technology adoption are probably wrong. The VHS lesson applies directly to generative AI tools, large language models, and automation platforms right now. The technology that wins will be the one your customers and employees trust and find accessible — not necessarily the most technically impressive one.
• A business case is not a vision statement. Module 6 taught me that an AI business case must address genuine organisational needs, measurable outcomes, and realistic implementation pathways. Vague AI ambition is not strategy. It is noise.

“Leadership in the AI age demands the courage to learn publicly, the discipline to go deep, and the humility to let the evidence change your mind.”

What Should We Do About It

Here are my direct, actionable recommendations for any leader serious about navigating the AI era:

1. Invest in structured education, not just awareness. Enrol in a programme like the Oxford Artificial Intelligence Programme. Commit the 7–10 hours per week. Do the assignments. Accept the feedback. This is non-negotiable if you want to lead transformation credibly.
2. Apply learning to your own organisation in real time. Every module I completed, I connected directly to my organisation’s context. This is not theoretical — it is operational preparation. Start doing this from week one.
3. Build ethical frameworks before you need them. Do not wait for a crisis to develop your AI ethics principles. Draft them now, test them against real scenarios, and embed them into your governance structure.
4. Challenge your AI vendors and internal teams. Once you understand the basics of machine learning and deep learning, you will start asking better questions. Better questions lead to better decisions and far fewer expensive mistakes.
5. Be honest about where your knowledge ends. One of the most valuable things rigorous study teaches you is the precise shape of your own ignorance. Know it. Name it. And then go close the gaps.

Frequently Asked Questions

Is the Oxford Artificial Intelligence Programme worth it for non-technical leaders?

Absolutely — and in many ways, it is designed for non-technical leaders. The programme builds conceptual understanding of machine learning, neural networks, and AI ethics without requiring a background in mathematics or programming. The emphasis is on strategic application: how do you identify AI opportunities, evaluate risks, and build a compelling business case? If you are a senior leader, a manager, or an entrepreneur, this programme gives you the vocabulary and frameworks to lead AI initiatives with genuine credibility.

How difficult is the Oxford AI Programme, and how much time does it require?

The programme requires approximately 7–10 hours per week across six modules. The difficulty is real — assessors provide substantive, critical feedback, and high scores require genuine engagement with the material, not surface-level responses. I scored between 56% and 100% across different assignments, which reflects the varying complexity of each module’s demands. Expect to be challenged. That challenge is exactly what makes the qualification meaningful.

What is the most important leadership lesson from studying AI at this level?

That responsibility cannot be delegated. You can delegate execution. You can delegate technical implementation. But the decision to deploy AI in your organisation — and the accountability for what it does to your customers, your employees, and your culture — sits with leadership. Understanding the technology is what makes responsible decision-making possible. Without that understanding, you are not leading. You are hoping.

The Bottom Line

Completing the Oxford Artificial Intelligence Programme with an 88.9% overall grade is something I am proud of — not because of the number, but because of what earning it required: genuine effort, intellectual honesty, and the willingness to be wrong in order to learn.

AI is the defining leadership challenge of our generation. It will not wait for you to feel ready. It will not slow down while you finish your current priorities. And it will absolutely not be led well by people who have never taken the time to truly understand it.

If you are a leader who is serious about navigating this era — not just surviving it, but shaping it — I would strongly encourage you to explore the Oxford Artificial Intelligence Programme for yourself. And if you have already taken a similar journey, I want to hear from you.

Drop a comment below: What has been your most important AI learning as a leader? Let’s build this conversation together.