Opening
I still remember the afternoon I first felt it properly. I had asked an AI tool to draft a short piece in my voice—nothing complex, just a few paragraphs. Seconds later, it produced something that sounded eerily like me. Not perfect, but close enough to make my stomach turn. For a moment I thought: if this can mimic my writing that fast, what happens to the people who do this for a living? What happens to me?
That quiet jolt of dread is what so many of us are feeling right now. It is not abstract worry about “the future of work”. It is the very real sense that something precious—our roles, our skills, our sense of usefulness—might slip away. Jobs that once felt secure suddenly look fragile. Identities built around what we do every day feel suddenly at risk. And beneath it all sits a deeper fear: that we are being replaced by something we do not fully understand.
A few months ago I wrote an essay trying to answer that fear with economics. I leaned on Ronald Coase and his idea of transaction costs: the hidden frictions that make some things too expensive to organise in the market. I argued that AI, by lowering those costs, would open up new kinds of work and ultimately create more jobs than it took away. The reasoning still holds. But looking back, I see the piece stayed too much in the realm of theory and logic. It spoke to the head, not the heart. It explained the mechanism without fully sitting with the emotion that makes people turn away from the argument in the first place.
This time I want to do it differently. I want to start where the fear actually lives—in the gut—and then walk with it towards something more hopeful. Because the shift we are entering will be disruptive. Painfully so for many. But it will also be full of real opportunity, if we choose to meet it with open eyes rather than clenched fists. The path is not easy, but it is not hopeless either. Let us begin by owning the fear honestly, then see what lies beyond it.
For me, this isn’t just an intellectual exercise. This summer, after 25 years of running two clothing shops with my wife Malin, we made the deliberate choice to close them down. Our two girls have partly fled the nest, and we could have kept going—trading comfortably, enjoying the rhythm we knew so well. But something deeper pulled at us. I felt an intuitive tug toward this emerging world of technology and AI. I wanted the freedom to explore it without the daily demands of the shops anchoring me in place. So we walked away. It was a leap, and it has brought its own mix of stress and uncertainty. Yet I have no regrets—only a forward gaze toward what might be possible.
Validating the Limbic Reality: Why the Fear Is Legitimate
That stomach-turning moment I described is not unique to me. It is the same quiet shock many people feel when they see AI handle tasks they thought belonged only to humans. A driver watches a video of an autonomous truck moving goods without a cab. A graphic designer sees software generate layouts that once took her hours. An accountant watches a tool summarise financial statements in seconds. Each one lands like a small personal earthquake. The fear is not irrational. It is the body’s honest response to a real threat.
We are wired to notice loss more sharply than gain. A missed opportunity stings less than something we already have being taken away. Picture the difference between finding a ten-pound note on the pavement and losing the same note from your wallet. The loss feels heavier, even though the amount is identical. This is not a quirk of character. It is how our minds protect what we value. When AI appears to threaten a job, it is not just income at risk. It is the daily rhythm, the respect from colleagues, the story we tell ourselves about who we are. Losing that feels like losing part of ourselves.
The pain is not only personal. It ripples outwards. A lorry driver who has driven the same routes for twenty years is not merely losing a wage. He is losing the quiet pride of knowing every lay-by, every shortcut, every story behind the wheel. A journalist who has built a career chasing leads suddenly sees writing tools produce articles that look passable. The craft she spent years mastering seems suddenly replaceable. Communities built around certain trades feel the tremor too. When a factory closes or an entire profession shrinks, the shops, the schools, the sense of belonging suffer.
Recent studies show that millions of jobs across developed countries carry a high risk of automation in the coming decade. White-collar roles once considered safe—legal research, basic coding, routine analysis—are now in the frame. The numbers are sobering. Yet the fear runs deeper than statistics. It is about the speed of change. Previous shifts, like the move from horses to cars, took generations. This one feels like it is happening in years. We worry that the safety nets we have built—retraining programmes, social support—will not keep pace.
I know this fear intimately. Closing the shops meant letting go of a world I had built over two and a half decades. The routines, the relationships with customers, the quiet satisfaction of seeing someone walk out happy with a purchase—they were all part of my identity. The decision felt right, but the transition has been disorienting. Some days the uncertainty weighs heavily; the path ahead is not yet clear. Yet that very discomfort has pushed me to lean into the questions this essay explores: What does this new landscape mean? How can we navigate it?
I am not saying the fear is the whole story. But it is the starting point. We cannot skip past it with promises of progress. If we pretend the unease does not exist, we lose the very people we need to help shape what comes next. The fear is legitimate because the stakes are real. And it is only by sitting with that reality that we can begin to see the possibility on the other side.
The Coasean Engine: Why Disruption Is Also the Door to Opportunity
We have faced this kind of fear before, and we have come out the other side with more than we lost. To see why, we need to look at what is actually happening beneath the surface. The key lies in a simple idea from an economist called Ronald Coase: the things we do in life are shaped by the costs of bringing them together.
Think of transaction costs as the everyday friction that makes some activities too expensive or too awkward to organise. Finding the right person for a job, agreeing on terms, checking that everyone keeps their word—these hidden costs often stop good ideas from happening. When those costs drop, suddenly a lot becomes possible that was not before.
The internet is the clearest example we have lived through. Before it arrived, many things were simply too hard to coordinate. If you wanted to buy a rare book from someone across the country, you had to search through catalogues, make phone calls, arrange payment and delivery. Most people did not bother. Travel agents had to call airlines and hotels one by one. Small businesses struggled to reach customers beyond their local area. The friction kept whole worlds of trade and connection locked away.
Then the internet came along and slashed those costs. Searching became instant. Payments moved online. Delivery could be tracked. Suddenly, people could buy from anyone, anywhere. Companies could reach customers they never could have before. New kinds of work appeared: web designers, digital marketers, online shop owners, freelance writers who could sell their words globally, app developers who built tools for this new connected world.
Yes, there was real pain. Travel agents lost their livelihoods. Print newspapers shrank. Video rental shops closed. Entire high streets felt the impact. Communities tied to those jobs suffered. The transition was slow, uneven, and for many people it was never easy. But the net result was clear: millions more jobs were created than were lost. Software engineers, content creators, e-commerce specialists, data analysts, remote customer support staff—the list grew far beyond what vanished.
AI is doing something similar, but to cognitive and coordination tasks rather than information and communication. It is lowering the cost of finding, matching, and organising knowledge and effort. A company that once had to hire a team to plan a complex project might now use AI to pull together data, draft plans, and monitor progress. Tasks that were too expensive to coordinate suddenly become feasible. New kinds of work emerge: people who oversee AI systems, who design prompts and workflows, who interpret results and make final decisions, who create entirely new services that were not viable before.
The disruption is real because old roles will shrink or disappear. But the opportunity is also real because the same mechanism that closes some doors opens many more. We are not witnessing the end of work. We are witnessing the removal of friction that has long held back human ingenuity. When that friction falls away, the economy does not shrink. It grows in ways we cannot always predict.
The Transition: Disruptive, Painful, and Full of Agency
The engine we have just described does not run on smooth roads. It runs on the rough ground of real human lives, and the journey is rarely comfortable. When transaction costs fall, old ways of working become less necessary. Some roles vanish quietly; others disappear in waves. The people who held those roles do not simply step into new ones overnight. They face months, sometimes years, of uncertainty, retraining, or even the need to start again in a different field. For many, the pain lingers long after the statistics show a net gain.
Consider the truck driver again who spent decades behind the wheel. His knowledge of roads, weather patterns, and the rhythm of long-haul life is deep and specific. If autonomous lorries take over, that expertise becomes less valuable in the market. He may find himself competing for jobs that demand entirely different skills—perhaps overseeing fleets of drones or managing logistics software. The shift is not impossible, but it is hard. It requires time, money, and emotional energy he may not have. Similar stories unfold for office workers whose routine tasks AI can now handle, or for creatives who see their craft replicated in seconds. The loss is not just financial. It is a blow to identity and purpose.
My own transition echoes this. Leaving behind the shops meant stepping into a space without the familiar structure I had known for so long. There have been stressful days—moments of doubt about what comes next, about whether I can find a meaningful path in this new territory. Yet those very moments of discomfort have become part of the exploration. By closing one chapter, I have gained the space to wander into AI and technology, to experiment, to ask questions, and to see what emerges.
The internet transition taught us this lesson clearly. Many who worked in travel agencies or print journalism never found their way back to the same level of security. Communities built around those trades felt the emptiness for years. Yet others adapted. A travel agent might have become a travel blogger or an online booking consultant. A journalist could shift to digital content strategy or podcast production. The new roles often paid better and offered more flexibility, but only for those who could bridge the gap.
The same pattern is likely with AI. The lorry driver could move into supervising automated delivery systems or analysing route data. A traditional accountant might become an AI-assisted financial strategist who focuses on client relationships and complex decisions. An artist could use AI as a collaborator to experiment at scales previously impossible. These paths exist, but they are not guaranteed. They require effort, support, and sometimes luck.
What matters most is that we are not passive in this process. We have agency. Individuals can learn new tools, take short courses, experiment with AI in their current work. Communities and governments can invest in retraining programmes, adjust education systems, and create safety nets that make the bridge easier to cross. The transition will be uneven—some will move quickly, others will struggle—but the door to opportunity remains open for those who choose to step through it.
We have seen this before. The pain is real. The possibility is equally real. The question is whether we meet the change with resistance or with the quiet determination to shape it.
Behavioural Science as the Hidden Map
The question of agency we just raised is not merely a matter of willpower. It is shaped by how our minds naturally respond to change. Behavioural science offers a quiet map through that territory, showing why the transition feels so heavy and how we can make it a little lighter.
Our minds have two speeds. One is fast, automatic, and emotional—always scanning for threats and sticking close to what feels familiar. The other is slower, deliberate, and effortful—able to weigh options and plan ahead, but it tires quickly and often lets the fast system take the wheel. When AI arrives, the fast system notices the threat first: lost jobs, changed routines, uncertainty. It reacts with alarm because change looks like loss. The slower system can see the longer picture—new roles, new possibilities—but it needs a moment to catch up.
This explains why the fear we feel is so stubborn. We are not ignoring the evidence of opportunity; our automatic mind is simply more tuned to danger than to hope. It is the same instinct that makes us grip the armrest during turbulence even when the pilot says everything is fine. The feeling is real, and it matters.
Yet the same science shows ways to shift the balance. Small nudges can help the deliberate mind step in. When we frame the change as a shared challenge rather than an individual burden, people feel less alone and more willing to act. When we start with one small experiment—trying an AI tool on a single task—the fast mind gets proof that the world does not collapse. That proof builds confidence, and confidence lets the slower mind plan further.
We also tend to overvalue what we already have. The job we know, the skills we have spent years building—they feel irreplaceable because they are ours. This makes letting go harder than it needs to be. But once we see others making the move successfully, the grip loosens. Stories of adaptation spread quietly, and what once felt impossible starts to look possible.
In my case, the act of closing the shops was itself a small (though large-feeling) experiment in letting go. It proved that the ground beneath me did not vanish. Instead, it opened up space for curiosity and new questions. The excitement of discovering what AI might enable—coupled with the honest fear of the unknown—has become my current frontier. It is thrilling and unsettling in equal measure.
In the end, the map is simple: acknowledge the fast mind’s alarm, give the slower mind room to breathe, and take small steps that prove the ground is solid. The transition will still be difficult. But understanding these patterns helps us meet it with clearer eyes and a steadier hand.
Closing: From Fear to Frontier
The map we have just walked through shows that fear is not the final word. It is the starting point. Our minds are built to notice threats, but they are also built to learn, adapt, and imagine what might come next. The choice we face is not whether AI will change things—it already is—but how we meet that change.
I am still working this out myself. Some days the uncertainty feels heavy. Other days I catch glimpses of what becomes possible when old frictions fall away: people spending less time on routine tasks and more time on the work that matters to them, new kinds of collaboration across distances, ideas that once stayed trapped in someone’s head now finding their way into the world faster than ever before. It is not a perfect picture. It is a more honest one.
What if we treated this moment like an expedition rather than a crisis? We would pack the fear—it is part of the luggage—but we would also pack curiosity. We would take small steps: try one AI tool on a task we already do, talk to someone who has started using it, share what we learn. We would remind ourselves that the internet once felt like a threat too, and yet here we are, connected in ways that would have seemed impossible thirty years ago.
The frontier is not without risk. Some paths will be steep. Some will lead to dead ends. But frontiers are where new things are found. They are where we discover capacities we did not know we had. For me, this journey began with closing one door and stepping into the unknown. It has been both exciting and fear-inducing, but every step forward feels like a quiet confirmation that there is more to discover.
I do not have the final answer for how this transition will unfold. I suspect no one does. But I know that when we meet change with honest fear and quiet determination, we tend to find our way forward together. That is enough to keep walking.
References
- Coase, Ronald H. (1937). “The Nature of the Firm.” Economica, 4(16), 386–405. https://www.jstor.org/stable/2626876 Seminal paper introducing transaction costs and the rationale for firms; forms the core economic mechanism in Section 3.
- Kahneman, Daniel (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux. https://us.macmillan.com/books/9780374533557/thinkingfastandslow Foundational work on System 1/System 2 thinking, loss aversion, and prospect theory; underpins Sections 2 and 5.
- Tversky, Amos & Kahneman, Daniel (1974). “Judgment under Uncertainty: Heuristics and Biases.” Science, 185(4157), 1124–1131. https://www.science.org/doi/10.1126/science.185.4157.1124 Classic paper on anchoring, availability, and representativeness heuristics; supports the discussion of how fear becomes sticky in Section 2.
- Thaler, Richard H. & Sunstein, Cass R. (2008/2021). Nudge: The Final Edition. Penguin Books. https://www.penguin.co.uk/books/111/111753/nudge/9780143137009.html Introduces choice architecture and libertarian paternalism; informs the agency and nudging ideas in Section 5.
- Ariely, Dan (2008). Predictably Irrational: The Hidden Forces That Shape Our Decisions. HarperCollins. https://www.harpercollins.com/products/predictably-irrational-dan-ariely Explores endowment effect, mental accounting, and zero-price bias; supports the discussion of why we overvalue existing jobs in Section 5.
- Autor, David H. (2015). “Why Are There Still So Many Jobs? The History and Future of Workplace Automation.” Journal of Economic Perspectives, 29(3), 3–30. https://www.aeaweb.org/articles?id=10.1257/jep.29.3.3 Balanced academic overview of how technology (including past waves like the internet) destroys and creates jobs; provides historical grounding for Sections 3 and 4.
- Brynjolfsson, Erik & McAfee, Andrew (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. W.W. Norton & Company. https://wwnorton.com/books/9780393350647 Accessible synthesis of how digital technologies (including AI) create net economic gains despite disruption; reinforces the optimistic but realistic framing in Sections 3 and 4.