Here’s a confession to start. I’m the de facto Chief AI Officer in my own business, and nobody gave me the title, the training, or the playbook. It sort of landed on me. I work with our head of research & operations, head of finance, and head of organisational development at Barden, who gave me an initial insight into what this AI stuff might do, so I got curious. My curiosity coincided with our team signing up with Claude and with the release of Opus 4.8, which changed the game entirely for me; but that’s a whole other story. So, for me, like for so many out there, I was not appointed; I just assumed the role.
Now I’m not pretending to be a technologist. I couldn’t write you a line of code if my life depended on it (though I could give a good go at some HTML from my college days) – but I run a firm that talks to leadership teams across the Irish market every week, and the same question keeps coming up – who is supposed to own this? So let me tell you what I’m seeing, and then let me argue with myself a bit, because the popular version of this story is just a little too clean-cut for my liking.
Why AI is not IT’s Job
When the internet arrived, we didn’t make it the CIO’s job to decide what the business should say on its website. IT kept the lights on. The business decided what to do with the lights. Somewhere along the way, we forgot that distinction with AI.
AI isn’t a system you install. It’s a way of working. It’s a tool that sits on the desk of the person in accounts payable, the person in marketing, the solicitor, the analyst, and it changes how their actual day goes.
That’s a business question. It’s a how does the work get done question. And IT, for all its brilliance, is not built to answer it. Ask IT to own AI , and you’ll get a beautifully governed platform that nobody uses, because it was never designed around how people actually work.
So, I’m with the popular view here. The value of AI is created in the business, by the business, for the business. Whoever owns it must live in that world.
Where does the Chief AI Officer come from?
My instinct says, not from a function at all. We need to look beyond the best technologist, the best operator, or the best strategist. The Chief AI Officer should be the person who understands how the whole thing hangs together; how a job moves from one desk to the next, where the friction is, where the boring bits live.
This job involves sitting down with the people who really know each corner of the business, the SMEs, the ones who do the work, and helping them imagine it differently. Not telling them. Imagining with them. What if the three hours you spend reformatting that report every week just… didn’t happen? What could you do with the three hours? Usually the more interesting work. The work people trained for. The work people actually enjoy.
That’s the heart of it for me.
To take the grind out of people’s weeks and hand them back the bits that make them want to do the job in the first place. You don’t need a computer science degree for that. You need to understand people, and you need to understand work.
Now Let Me Argue with Myself…
That’s the romantic version. Here’s where I think the popular take and, if I’m honest, my own first draft of it, is a bit wrong.
First, “it’s not IT’s job” is half right and half lazy. The business should own AI, yes. But the idea that IT just steps aside is dangerous. Data quality, security, what’s allowed near client information, what happens when AI confidently makes something up – that’s not optional, and it’s not the business’s home turf.
Realistically, the honest model isn’t “business instead of IT”. It’s “business leads, IT makes it safe”. A Chief AI Officer who treats IT as an inconvenience will get the firm into trouble fast. Shadow AI, where people quietly paste sensitive stuff into random tools, is already a live risk in plenty of organisations. That problem only gets solved if both the business and IT are at the table to troubleshoot.
Second, “comes from the business, not a function” sounds great until you need to actually hire one. In the real world, the people doing this job well tend to come from somewhere – transformation, analytics, operations, strategy. Pure business fluency with zero technical literacy is its own trap. If you can’t tell the difference between what these tools genuinely do and what a slick vendor says they do, you’ll get sold a pup. So, I’d soften my own line:
Third, and this is the one nobody likes saying out loud – “removing the mundane to free people up for interesting work” is sometimes true and sometimes a euphemism. Freed-up time can become better work. It can also become fewer people. Any leader pretending that tension doesn’t exist isn’t being entirely honest. The Chief AI Officer who’s worth having is the one willing to have the harder conversation about what the work, and the workforce, actually looks like on the other side of this, not just the nice slide about reclaimed hours.
The Bit I Keep Coming Back To
So what is the role, really? It’s a translator. Someone bilingual; fluent in how the business breathes and literate enough in the technology to call nonsense when they hear it. Someone who can sit beside an SME, help them picture a better Tuesday, and then work with IT to build it without blowing a hole in the firm’s security. Ultimately, AI is just another tool to help us improve the efficiencies in what we do and get to do things that we never had the time to do before.
“In my experience, successful AI implementation has far less to do with choosing the right model than understanding the business well enough to know where AI should and shouldn’t be applied. The organisations creating the most value are redesigning workflows around better customer outcomes, stronger decision-making and improved employee experience, while working closely with technical teams to ensure those changes are secure, governed and scalable. AI leadership is ultimately a business transformation challenge enabled by technology, not the other way around.” – Alex Kennedy (Profile here >>>) Independent Advisor – Commercial Operations, AI Implementation & Revenue Systems
Final Thoughts
The Chief Electricity Officer was a real job once. When power was new and strange, big firms appointed someone senior to figure out what to do with it. You don’t have one now because electricity stopped being a project and became just… how things work.
The best Chief AI Officer might be the one who subtly makes their own role unnecessary. Who spreads the fluency so wide that, in a few years, nobody needs a title for it at all.
That, for me, is the job worth doing.
Ed Heffernan is the Managing Partner of Barden Ireland. Connect with Ed at ed.heffernan@barden.ie or on LinkedIn>>>
