Human-Centred Intelligence: Leading IT in the Age of AI
A practical guide for IT leaders navigating the next era of intelligence.
The pressure to “deliver AI” has become the new corporate mantra, but what does it really mean?
I’ve practically lost count of the times I’ve sat in front of peers, industry leaders, interviewers, merger and acquisition teams, private equity, and a whole host of business leaders, managers and CEOs who have all asked me for the same thing.
“We want you to deliver AI.”
I’m positive that IT leaders across the globe are consistently hearing the same request, and many of us are working hard to suppress a blank stare while desperately piecing together the right words to sound visionary, strategic, and in control. Interviewers expect something plausible and innovative. Executives want a clear roadmap. The business demands tangible value. And yet, that’s often difficult to produce for three main reasons:
Business leaders often ask for “AI” without a clear definition of what outcome they expect.
IT leaders are still defining what “delivering AI” looks like in practice.
Strategic alignment is unclear; many teams struggle to turn ambition into a practical, value‑driven plan.
Sometimes, it’s a blend of all three, resulting in awkward conversations and the occasional tactical side-step to avoid eager CEOs while managing the very real frustrations of modern tech leadership. I’ve seen talented IT leaders get it completely, hopelessly wrong, not out of incompetence (occasionally out of incompetence), but out of the pressure to sound certain, and be certain, in an uncertain space.
So where does that leave today’s IT leader in the age of AI? When everyone around you is looking for you to understand, looking to you for all the answers, how do we move forward confidently as IT leaders? How do we effectively leverage the AI landscape and deliver on its promises? Creating the workplace of the future is not all as clear as we’d like it to be.
In this article I want to cover some of these concepts, this will help set some of the foundations for future writings, but will also clarify some of our immediate challenges as IT tech leaders. Today we’ll cover:
The IT Leader’s New Frontier, an introduction to how AI is reshaping the identity and expectations of IT leadership.
The Shift From Ownership to Orchestration, a look at why modern IT leaders must guide ecosystems rather than own every system.
The New Skillset of the IT Leader, a breakdown of the evolving capabilities needed to lead confidently in the AI era.
Building the Foundations Before the Tools, a reminder that without solid data, process, and governance foundations, AI will fail before it begins.
The Danger of AI Theatre, an honest look at superficial AI initiatives that create noise instead of value.
AI as Cultural Transformation, an exploration of how mindset, trust, and behaviour determine AI success more than tooling.
The Critical Role of Problem Definition, a dive into why defining the right problem matters more than choosing the right model.
Aligning AI with Real Business Value, a look at tying AI initiatives to meaningful, measurable outcomes.
Moving from Firefighting to Foresight, exploring how AI elevates IT from reactive support to proactive strategic leadership.
Why “I Don’t Know” is a Leadership Superpower, a reflection on how honesty and curiosity build stronger, more credible leadership.
1. The IT Leader’s New Frontier
AI is transforming IT from a business support function into a proactive engine of business productivity and intelligence. But the tools alone don’t define success; the people who use them do. The new IT frontier is about orchestrating humans and machines toward shared goals. It’s about empathy in automation and foresight in data. It is not about purchasing an AI licence and hoping it magically transforms the organisation. It’s not about panic‑buying every tool with an AI sticker on the box. It’s not swapping out mature, reliable systems for overpriced “AI‑powered” equivalents that introduce more complexity than value. Simply put, tool shopping is not a strategy. Buying licences is not a transformation. Real AI adoption requires clarity, capability building, problem definition and a deliberate plan that connects technology to meaningful outcomes. A simple fact that is incredibly lost on most, myself included, when the AI revolution started building momentum.
We’ll cover some of this in greater depth, but in short, the role of the IT leader will change rapidly over the next few years, perhaps even the next months, what with this rapid rate of change. It’s fascinating to be part of the next technical revolution, and current and future IT leaders will be the ones leading the charge and guiding others and businesses to help them succeed.
2. The Shift From Ownership to Orchestration
Critical to this success will be the shift from ownership to orchestration, foundational in fact. The cloud has quietly rewritten the rules of IT, and AI has made that shift impossible to ignore. Most modern AI capability lives in the cloud, built on infrastructure, models, and compute power that no individual organisation could realistically own. This means IT leaders now operate in an environment where the majority of the technology stack sits outside their walls.
Instead of owning systems end‑to‑end, IT now orchestrates an ecosystem of cloud services, vendors, APIs, automations, and data flows. The role has evolved from builder to conductor, from controlling every component to designing how the components work together.
This shift is driven by reality:
Vendors now innovate faster than internal teams can track.
Data flows across dozens of platforms, not a single controlled environment.
Business teams adopt tools independently, forcing IT to guide rather than gatekeep.
AI services are consumed rather than installed.
Ownership required certainty. Orchestration requires curiosity, clarity, influence, and systems thinking. The shift has been happening for a while, but those of us still holding onto infrastructure ownership will need to learn to start letting go. It’s no longer about controlling and holding everything tightly but about ensuring we can manage hybrid environments, ensure everything works together and then ultimately prepare for a shift to a fully cloud-managed environment.
3. The New Skillset of the IT Leader
The modern IT leader’s skillset is shifting from technical mastery to strategic intelligence, the ability to understand how technology, data, people, and business goals intersect. You could argue and say this is what a lot of IT leaders do, particularly those holding a CTO badge, but all IT leaders are feeling a shift as this responsibility starts to move further down the middle management chain. So, because of this shift, take note of the following:
Be adaptive and comfortable with change
An ability to embrace change and remain adaptive will be essential in the current and future AI workplace. This ties closely to the rapid pace of technological evolution. Adaptability keeps you relevant. AI itself is the new skill of the decade, not just its tools, but the underlying technologies that power it, particularly data science, automation, and modern programming practices. Comfort and understanding with these technologies will be critical.
Translating business needs into data opportunities
The most valuable leaders can articulate how data supports decisions, improves outcomes, and unlocks opportunities. They bring clarity where the business brings complexity, and help teams understand how AI can transform that clarity into capability.
Ethical awareness and human‑centred leadership
AI also introduces new risks, biases, and uncertainties, which have inevitably freaked many people out. Outlandish claims of AI stealing jobs and taking over the world by key public figures have left many feeling uneasy. This leaves it to us to set the tone by ensuring technology is deployed responsibly, transparently, and in service of people, not at their expense. A clear ethical framework will serve all leaders of course but the specifics of ethical AI carry a significant amount of weight at the moment. People are looking for clear, moral guidance. Human-centred AI is an important ethical perspective.
4. Building the Foundations Before the Tools
Some of my predecessors went straight to the tools in eager adoption of AI, but there is so much to consider operationally, so much so that without the groundwork covered, even those tools will be shaky at best. This creates immediate doubt culturally and professionally, so it’s best to get our AI ducks lined up before we start diving in.
Data quality and accessibility
AI for all the intelligence it delivers (or seems to deliver) is a result of having the right data. AI cannot compensate for missing, incorrect, or inconsistent data, and most businesses have a data landscape to be feared, at the very least, looked at sceptically. IT leaders must understand where their data lives, how clean it is, and how easily it can be accessed. Without this, every AI initiative becomes guesswork dressed up as progress.
Process clarity and standardisation
I’ve had the pleasure of working in a number of different industries, businesses of varying sizes and capabilities but this is always an area of struggle or at least refinement. In most cases, processes are undocumented, inconsistent, or vary wildly across teams and because of this, AI has nothing stable to latch onto. Clear processes create predictable patterns, which help the AI adapt and learn quickly. Not only that but clear process documentation will assist with the issues around problem definition which I describe later.
Integration and interoperability
Sadly, businesses are often a patchwork of legacy systems, modern apps, spreadsheets, and shadow IT. AI thrives in connected environments, and rarely do we have the ability to ensure systems can talk to each other, a prerequisite to business intelligence and AI. It’s often we find that any AI strategy worth the computer it’s typed on will require plans to out some if not all of these legacy platforms. Aging ERP, HR, and ropey telephony platforms are often some of the first targets.
Governance and security
An important consideration, without any doubt and often deployed by sales companies to drive fear into AI adopters. How often I’ve heard the story of a University inadvertently sharing payroll data to public LLM (Large Language Model) is unreal but it paints a clear warning. AI introduces new data flows, permissions, and compliance considerations. Without governance, leaders risk deploying solutions that expose data, breach regulations, and ultimately undermine the trust of AI. Do your homework, get this right, get an outside consultant if needed to plug this potential governance and security gap. Better safe than sorry.
Capability building and workforce readiness
Tools will not save a team that isn’t prepared to use them. Training, experimentation time, and skill‑building are essential. As IT leaders, we’re the voice to give people confidence, but ensuring that we’re fully prepared prior is critically the most important thing. If we’re not confident and clear in what we’re trying to deliver, our teams nor the business will follow suit. Get comfortable and clear on your strategy before you start shouting it from the office roof.
Why this matters
Most organisations want to jump straight to tools because tools feel like progress. But without these foundations, AI becomes slow, unpredictable, inaccurate, or politically unworkable. The smartest leaders know that the work no one sees is the work that makes everything else possible.
5. The Danger of AI Theatre
I may have accidentally raised the matter of incompetence earlier and this is what I was specifically referring to, AI Theatre. It’s the performance of progress without the substance of progress. It’s the organisational habit of running flashy pilots, producing polished slide decks, or name‑dropping AI models without ever delivering meaningful value.
A real example of this was the delivery of Microsoft Teams Transcription, long before the days of meeting summaries, auto-built task lists and emailing following up actions. Just a transcript. This AI breakthrough was touted as the next business breakthrough, every manager and business leader got pulled into meetings, was shown a flashy slide deck and then it was demonstrated.
Anyone familiar with early or even todays raw transcriptions will know how painful they are to read and decipher, on their own they deliver very little but the theatre in which it was presented built it up to be so much more, only to then watch the reputation of AI and the individuals involved come crashing down. This undoubtedly stalled AI adoption at the business.
They were trying to solve a problem that didn’t exist. It momentarily created the illusion of innovation while masking the absence of any meaningful strategy. The danger wasn’t just wasted money and time; it’s the erosion of trust in AI itself. When executives see pilots go nowhere, they start to believe the technology has no real impact, when in reality, it was never given the structure or sponsorship it needed. The IT leader’s role is to recognise when AI is being performed rather than implemented, and redirect effort toward problems that matter.
6. AI as Cultural Transformation
Tech adoption and particularly AI adoption, fail far more often because of people than because of technology. The tools are ready. The real question is whether the organisation is.
True AI transformation requires a shift in mindset: curiosity over caution, experimentation over perfection, and learning over defensiveness. Teams need psychological safety to try new approaches without fear of embarrassment, and all leaders (not just IT leadership) must create the conditions for that safety. Without it, AI becomes something people avoid rather than embrace.
Many organisations rush to deploy tooling because it feels like visible progress. But deploying a tool without preparing the culture almost guarantees resistance. People don’t reject AI because it’s complicated; they reject it because it threatens routines, comfort, identity, or control. Cultural readiness is what bridges that gap.
Business culture also doesn’t move at the speed of code or cloud innovation. It moves slowly, through trust, clarity, communication, and shared understanding. That’s why tech leaders must act as cultural translators, helping teams understand not just how AI works, but why it matters and what it enables.
Cultural transformation requires cross-functional partnership. HR, operations, finance, IT, and executive leadership all shape how people experience change. If even one of those functions drags its feet, adoption slows to a crawl.
The IT leader’s job isn’t to force AI onto the organisation. It’s to guide people through the uncertainty, build alignment, and create a culture where intelligence, human and machine, can thrive.
7. The Critical Role of Problem Definition
Problem definition is where most AI ambitions either take shape or fall apart, and in my experience, this is the bit everyone rushes past because it isn’t shiny, it isn’t fast, and it doesn’t feel like “AI adoption”. But this is exactly where the real work begins.
A clear problem definition is the single greatest predictor of whether an AI initiative will succeed. Most organisations begin with a solution in mind, “We need AI for forecasting”, “We need AI for stock control”, “We need AI to improve customer experience”, when in reality, these are outcomes, not problems.
When someone says “We need AI for forecasting”, what they really mean is “We don’t fully understand why our forecasting isn’t working”. When someone says “We need AI for stock control”, the real issue is “We don’t know where the friction is, who owns the data, or why the process keeps breaking”. That’s the level we need to get to.
The job of an IT leader isn’t to force AI into a business problem; it’s to keep asking why the problem exists in the first place. Sometimes AI is the answer. Sometimes it isn’t. And sometimes the hardest part is admitting that a broken process or a lack of data maturity is the real culprit, not the absence of a model.
This is also where honesty matters. Buying CoPilot or rolling out ChatGPT isn’t going to magically turn a business into an AI‑driven organisation. These tools can support intelligence, but they cannot create it from thin air. If we don’t understand the problem deeply, AI becomes a guessing game, and guessing is expensive.
Strong problem definition keeps us disciplined. It keeps us honest. And it ensures that when we do bring AI into the equation, it’s solving something real, measurable, and worth solving.
8. Aligning AI With Real Business Value
If there’s one thing AI has done brilliantly, it’s convince every business that “value” automatically appears the moment you plug in a model. We know better. AI only creates value when the problems are clear, the processes are stable, the people are ready, and the outcomes are measurable. Everything else is noise.
Aligning AI with real business value starts with a blunt question: What are we actually trying to improve? Not theoretically, not aspirationally, but in the day‑to‑day reality of the business.
Sometimes the answer is efficiency. Sometimes it’s accuracy. Sometimes it’s saving people from spending their lives in spreadsheets. Sometimes it’s unlocking insights that the business never had access to. Whatever the case, the value must be specific, visible, and understood.
One thing I’ve learned the hard way is that value isn’t created at the moment of adoption; it’s created at the moment of integration. When AI is woven tightly into the workflow, when it changes how decisions are made, when it nudges behaviour in better directions, that’s where the magic happens.
The challenge is that many organisations tie AI to value after the work is done, retrofitting benefits that don’t truly exist. As leaders, we have to resist that temptation. We define the value before the project starts, we track it as the work unfolds, and we validate it once deployed. That discipline stops AI from becoming theatre and keeps it anchored to outcomes that matter.
Most importantly, aligning AI with value isn’t about ROI spreadsheets or consultancy frameworks, it’s about asking whether the people using it feel supported, empowered, faster, clearer, or more capable than they did before. Real value is felt long before it’s measured.
When we get this right, AI stops being a science project and becomes what it should have been all along: a meaningful contributor to how the business thrives.
9. Moving From Firefighting to Foresight
If there’s one experience every IT leader shares, it’s living in constant firefighting mode. Tickets flying in, systems wobbling, users frustrated, and every day feeling like a sprint through an obstacle course. For years, this has been the default setting of IT, reactive, overloaded, heroic, and exhausted.
AI finally gives us a way out of that cycle.
The real promise of AI in IT isn’t automation for automation’s sake; it’s the shift from responding to anticipating. From waiting for something to break to knowing it’s going to break long before anyone else does. From juggling escalations to preventing them quietly in the background.
This is what foresight looks like:
spotting patterns in performance data long before thresholds are crossed
predicting user needs before they become pain points
identifying bottlenecks in workflows before they slow the business down
uncovering improvements that no human had the time or visibility to notice
And here’s the subtle but important part: foresight changes how IT is perceived. When you move from break–fix firefighting to strategic insight, you stop being the department that “fixes things” and start being the function that guides the future of the business.
AI won’t eliminate every fire, but it will shrink the forest. It gives leaders the fuel to move away from operational chaos and toward intentional, proactive strategy, the kind of strategy that not only improves uptime but also elevates IT into the realm of genuine business leadership.
Foresight is another area where the value of AI becomes visible, tangible, and deeply respected. It’s where IT stops reacting to the business and starts shaping it.
10. Why “I Don’t Know” Is a Leadership Superpower
There’s a strange pressure in tech leadership, this idea that the moment a new wave of innovation arrives, we’re supposed to become instant experts, fluent in every model, every acronym, every opportunity it presents. It’s nonsense, of course, but it’s amazing how quickly that pressure convinces smart people to bluff.
The truth is simple: the leaders who try to look certain are the ones who make the most damaging mistakes.
Saying “I don’t know” isn’t an admission of incompetence. It’s a statement of responsibility. It tells the room that you value accuracy over ego, progress over performance, truth over theatre. And in a landscape as fast-moving as AI, honesty is one of the most stabilising forces a team can have.
When you say “I don’t know yet, but I will”, you’re doing three powerful things:
setting a realistic pace for the business
creating space for learning and experimentation
anchoring the organisation in evidence, not assumptions
People trust leaders who tell them the truth, especially when the truth is uncomfortable. Pretending to know everything in AI isn’t leadership; it’s gambling with the credibility of the entire tech function.
Real leadership is steady, grounded, and unafraid to acknowledge uncertainty. And ironically, the leaders most willing to say “I don’t know” are often the ones who end up knowing more than anyone else.
If you’ve made it this far, you’ll know that AI isn’t a magic switch, and it certainly isn’t a shortcut. It’s the next evolution of how we lead, how we think, and how we build organisations that are genuinely ready for the future. Not because we bought the right tool, but because we built the right foundations, asked the right questions, and led with honesty rather than bravado.
The role of the IT leader is shifting faster than any time in recent memory. We’re no longer the people who simply keep the lights on. We’re the translators, the orchestrators, the cultural shapers, the problem definers, and the steady hands guiding businesses through a landscape that changes by the week.
This isn’t easy work, but it is meaningful work. And if we approach AI with clarity, integrity, and a human‑centred mindset, we don’t just adopt new technologies, we elevate the entire organisation.
So take this as your reminder: you don’t need all the answers, you don’t need perfection, and you certainly don’t need to pretend. What you need is curiosity, courage, and the willingness to lead from a place of truth. The rest will follow.
This journey is only just beginning for all of us. And if we do it right, it’s a journey that will redefine what intelligent leadership looks like for the next decade and beyond.