Staying human on purpose: leading in the age of intelligent machines

Yesterday I had the privilege of giving a keynote at the #UCQlearningfair2026 in Leeds, speaking to around 150 work-based learners and apprentices. My thanks to Michelle Elliott (Vice Principal) for the invitation and to Dr Jenny Davidson for the introduction. I got to know UCQ through a research project with Jenny, looking for the secret behind its remarkable student satisfaction ratings, and it was great to physically connect with that community. This post is a written version of what I said.
I started with two questions, asking for a show of hands. First: who is confident they know what their job will look like in five years' time? Very few hands. Second: who expects to be leading other people through that uncertainty for the next five years? Almost everyone.
That gap is what I wanted to talk about - what it means to be the human in the middle of all this change, leading other humans. That is what leadership has always had to be, but it is harder, and more important, than ever before.
A little context on where I am coming from. I have led people for 34 years - from shift workers on packing lines in a shampoo factory at Procter & Gamble, to teams of academics and professional services staff as MBA Director at Newcastle University Business School. I also got my PhD at 50 with two children in primary school, so I have some sense of what it means to work, study and juggle a million other things at once.
When the source of disruption can think
Back in 2017, the McKinsey Global Institute estimated that around 375 million workers globally may need to change the nature of their work by 2030 because of automation and AI, and its 2026 research suggests AI could already take on more than half of working hours. That doesn't necessarily mean humans will be doing less. More likely the pie is getting bigger, and both humans and machines will be doing more.
It's hard to know what to do with this kind of information. It's a bit exciting and a bit concerning - and the people we lead are likely to have mixed feelings too.
Most of us have had some training, formal or informal, in leading in what the US Army War College first called a VUCA world: volatile, uncertain, complex and ambiguous. We are starting to appreciate that this is not a passing phase. As leaders we still have to stand in the middle of it, offer a vision of the future, and help people to do what needs to be done to get there.
What feels different from earlier waves of rapid change is that, for the first time, the source of the disruption can think. Not like us - I'm not convinced AI has consciousness - but it can do many of the things we do, much faster, with enormous pattern-recognition capability, and without ever having a bad day or needing to stop for lunch.
There is a warning here too. Researchers at the Wharton School studied what happens when people work alongside AI and described something they called cognitive surrender: the moment when, instead of thinking critically about what AI tells us, we simply accept it. When the AI was accurate, people's performance rose by 25 percent. When it was wrong, their performance dropped 15 percent below what it had been without AI. In 73 percent of cases where the AI was wrong, people did not question it.
They were not less capable. They had handed their judgement to the machine. AI can suppress our own thinking and make us feel more confident at the same time - a dangerous combination. It's up to all of us to hold on to our confidence in our own judgement.
The gap where the tea gets made
Jonathan Boymal, an applied economist at RMIT in Melbourne, writes beautifully about this on LinkedIn and in his Substack, The Last Analogue. Earlier this year he wrote:
"Humans communicate at roughly two thousand bits per minute. The machines we have built exchange information millions of times faster. By any computational measure, we are spectacularly inefficient. Efficiency, however, is not the same thing as meaning."
And later in the same piece:
"The fear we keep rehearsing - that machines will become too human - misunderstands the problem. The challenge is that they will never be human enough. And that we will act, for reasons of cost and convenience, as though the gap does not matter. It matters because the gap is not empty. It is where the tea gets made. Where the failing student discovers someone noticed. Where justice becomes something more than procedure."
The risk is that we start to behave as if the gap between us and the machines isn't there. But that gap is exactly where leaders need to step in. We need to know when a walk around the block or a cup of tea is what's needed. You can't matter to a large language model. You can matter to a human.
Many thinkers have tried to put into words what is special about being human. The philosopher Emmanuel Levinas argued that the face of another person is the origin of all ethics - before any rules or governance framework comes the encounter with another human being and their claim on us. Martin Buber called it the I-Thou encounter, when we treat another not as a means to an end but as a full person. The sociologist Hartmut Rosa distinguishes resonance - being genuinely touched and changed by another person - from echo, where the other simply reflects back what we send. Algorithmic systems, he suggests, have become extraordinarily sophisticated echo chambers.
Most of us work in organisations where everyone is said to matter. But does it really feel like that? And what is our role in making it true? It's worth asking what does it for you. Does a personalised notification make you feel you matter? How do you know - in your body, not just your inbox - that you were actually seen?
Research from this year on teams using generative AI found that as AI takes on more of the technical work, human leadership matters more, not less. Coaching, empathy, psychological safety and making people feel genuinely seen remain irreplaceable. When machines can do more of everything else, a leader's capacity to make someone feel they matter becomes their distinctive contribution.
We feel we matter when another human pays attention to us. Simone Weil wrote that "attention is the rarest and purest form of generosity". Giving someone your full attention - not half-watching a screen or glancing at your watch - means noticing their tone of voice, sensing what isn't being said, and asking the question nobody else thought of or was brave enough to ask. No machine can match that, and it is at the core of what leadership has always been at its best.
Staying human on purpose
I am not arguing against AI. I get a real thrill from what it can do, I use it every day, and I still feel like a novice in the foothills of learning how to use it well. The question is not whether to use it but how - and using it well takes self-control and very conscious choices.
The Wharton researchers distinguish cognitive surrender from cognitive offloading. Offloading is using AI as a strategic tool while keeping your own judgement engaged - like using a calculator while still understanding the maths. You remain the author. The trouble is that the tipping point between the two is almost imperceptible.
Ethan Mollick, whose Substack One Useful Thing is also well worth following, argues that the most important thing we can do is be intentional about what we hand over and what we keep. It's not a matter of rejecting AI or adopting it uncritically, but of making deliberate, ongoing choices about what we use it for and when we stop.
There is a deeper reason to take care. Economists describe the Grossman-Stiglitz paradox: if markets provide all information freely, nobody invests in discovering it. Applied to AI, if the machine gives us all the answers, we may stop developing the capacity to find them ourselves. The muscle for independent critical thinking could weaken - and in younger people who have never worked without AI, it may never fully develop. Yet that independent judgement is what lets us read a room and know what is right for these particular people in this particular situation.
Staying human, then, has to be done on purpose. It's a professional necessity. And it isn't easy. Gary Turk's spoken-word poem Look Up went viral as smartphones were catching on, and its question still stands: what are we showing the people around us when we reach for our phone instead of being available for human interaction? When we are scrolling we are not just distracted - we are less available for the kind of connection that makes people feel they matter. Being the one who looks up, when everyone around you is looking down, is a choice that has to be made over and over again.
Leadership happens at every level of an organisation, and it rests more on character and practical wisdom than on any technical competency. It can be developed through practice, reflection and good role models. And whether or not you feel like one, you are a role model for the people you lead - even when you are making a mess of things, perhaps especially then.
Three questions and one piece of homework
I left the audience with three questions - not for answering on the spot, but for the journey home, a conversation with someone you trust, or reflection in a journal.
When I give AI a task and it does the thinking, what am I bringing, what am I gaining and what am I losing? What judgement, context or care do I add that the machine cannot - and what do I lose by not doing it myself?
How can I make sure everyone on my team feels they matter?
How can I make it easier for my team to think, rather than to stop thinking? Am I more likely to ask questions or give answers? Do I make space for challenge, and help people feel confident to bring their own judgement?
And the homework: have a conversation with another human being about something that matters to both of you, and give them your full attention.
Whatever stage of your leadership journey you are at, and whatever the uncertainty ahead, what is needed in practice is you - not a machine, a programme or an agent, but a real person. Someone who makes other people feel they matter, and who keeps thinking even when the chatbot produces an answer before you've finished typing the question. In a world where so much is being automated, that irreplaceable, inefficient, bandwidth-limited, spectacularly human thing you are, is worth protecting.
Further reading
Boymal, J. (2026). The Bandwidth of Being. The Last Analogue (Substack).
Shaw, J. and Nave, G. (2026). Thinking Fast, Slow, and Artificial. Wharton School.
Mollick, E. (2026). Choosing to Stay Human. One Useful Thing (Substack).
Ergun et al. (2026). Leading GenAI-Equipped Teams.
McKinsey Global Institute (2017). Jobs Lost, Jobs Gained; and (2026). Agents, Robots, and Us.
Brooks, D. (2021). A New Generation of Wise Thinkers and Good Leaders. Oxford.
Rosa, H. Resonance: A Sociology of Our Relationship to the World.
Buber, M. (1923). I and Thou.
Weil, S. Waiting on God.
Turk, G. Look Up (spoken-word poem).



Comments