Three years ago, I was put in front of Wall Street analysts as an "AI expert". I remember feeling like a bit of an imposter; surely there were people who knew far more than me. But what I found, again and again, was that I knew more than most of the room needed me to. Not because I was some oracle of machine learning, but because I'd spent time with the tools and thought seriously about what they meant for how businesses actually work.
That gap between "all-knowing" and "knows enough to be useful" has stuck with me ever since.
In the years since, the landscape has changed enormously. A huge number of people and businesses have made AI their entire professional focus. Just this week, I enjoyed a presentation by Emma from Spark showing how they are leading the way in helping advertising agencies adopt AI meaningfully. Alongside them, a wave of startups has emerged in areas like proptech and healthtech, where vertical-specific rich datasets and clear use cases have made AI adoption almost inevitable.
I would argue all of this has been positive value-adding innovation, and reports of the contribution of AI to the UK economy bear this out. But it's also raised a question I keep coming back to: does every senior leader now need a working knowledge of AI, regardless of whether they run a tech company, an advertising agency, or anything else entirely?
I think the answer is yes. Not expert-level. Not "can build a model" level. But a real, working fluency: enough to ask the right questions, spot the difference between genuine capability and hype, and make sound calls about where AI actually changes the shape of a business.
I'd go further: this should now be a fundamental criterion when a business is choosing who leads it, in any sector.
This week I was also lucky enough to sit in a room full of women leaders discussing AI. It was clear that the levels of understanding and adoption were higher than average, but still varied significantly. So what actually counts as "good enough" when it comes to a leader's understanding of AI? Is it being able to hold your own in a conversation with your CTO? Is it having used the tools yourself, hands-on, rather than just commissioning a strategy deck about them? Is it having a clear position on the ethics of its use, or indeed something else entirely?
I don't have a tidy answer. But I think it's a question every board, every search committee, and every leader evaluating themselves should be asking right now.