Three Narratives for the Future of Work
Mass Displacement, Leap of Consciousness and Managed Deceleration
An essay by Paulo Soeiro de Carvalho
Setting the Stage
A widely cited figure from the World Economic Forum’s Future of Jobs Report 2025 is that by 2030 the global economy could see around 170 million new roles created and around 92 million displaced, implying a net increase of roughly 78 million jobs. It is an attractive number, and it fits neatly into the story we like to tell ourselves: technology disrupts, society adapts, progress continues.
But there is a dangerous illusion hiding inside that “positive balance.”
Because the future of work is not decided by net totals. It is decided by timing, distribution, transition capacity, and the social contract that connects people to value creation. A labour market can show a positive balance and still produce instability—if the jobs created do not match the jobs destroyed, if they appear too late, if whole cohorts are left behind, or if the mechanisms that transmit wealth through work begin to fail.
That is why, when asked whether I am optimistic or worried about the future of work, my answer is deliberately uncomfortable: I refuse the binary. I do not think we should be “optimists” or “pessimists.” We should be prepared.
Preparedness requires a different way of thinking: not forecasting a single future, but expanding the space of possible futures—and building strategies that remain viable across multiple outcomes. In foresight, we do not ask “what will happen?” as if the future were a destination. We ask: what could happen, under which conditions, and what would we do if it did?
That is the purpose of the three narratives in this essay.
Introduction: why narratives are more useful than forecasts
When we talk about the future of work, we often do it in the language of prediction: how many jobs will be replaced, which occupations will survive, what the labour market will look like in 2040. Yet the real value of futures thinking is not in pretending we can forecast a single outcome with precision. It is in strengthening our ability to think strategically under uncertainty.
That is why I prefer to frame the future of work through narratives: coherent, plausible logics of change that help us rehearse different types of disruption. Narratives are not “stories” in the superficial sense. They are structured explanations of how a system could evolve—economically, politically, socially, and psychologically—given the same underlying drivers: AI, automation, robotics, platform economics, geopolitics, demographic pressure, and shifting expectations of meaning and identity.
The three narratives below—Mass Displacement, Leap of Consciousness, and Managed Deceleration—are not mutually exclusive. Elements of all three may co-exist in different sectors, countries, or time windows. But treating them as distinct futures is useful because each one forces a different set of questions, and each one reveals different vulnerabilities in our institutions, economies, and personal life-design assumptions.
Before we enter them, one contextual point matters.
The deeper context: a disruptive revolution, not a normal cycle
We are not dealing with “just another” technological change. We are dealing with a disruptive transformation driven by AI—particularly neural-network-based systems—expanding across digital, virtual, and physical spaces. Language models, image models, video models, and agentic systems are not only automating tasks; they are building a new layer of capability that begins to touch the very centre of white-collar work: the manipulation of information, data, knowledge, and decisions.
A large share of cognitive execution—administration, processing, routine analysis, reporting, support functions—will increasingly be performed by machines. The open variable is not direction. The open variable is speed. And speed is the difference between a manageable transition and a systemic rupture.
Narrative 1 — The Revolution of the Irrelevant
When job destruction outpaces society’s capacity to adapt
The first narrative begins with a blunt premise: this technological revolution—especially the transformation associated with artificial intelligence—will eliminate an extremely significant number of jobs without creating alternative jobs at the scale required to replace them. The disruption is not simply “automation happens.” It is that automation happens faster than the system can reconfigure.
To understand why this becomes difficult to manage, we do not need a science-fiction scenario of “no work.” We need only a shift in the structural level of unemployment that advanced economies can absorb. If a country accustomed to unemployment rates around 4–6% moves to sustained levels of 10%, 15%, or even 20%, the system behaves differently. Labour markets stop functioning as integration mechanisms. Public finances come under strain. Social mobility weakens. And the probability of political disruption rises—not because collapse is inevitable, but because the legitimacy of many democratic systems depends on broad participation in economic life.
In this narrative, the critical question is not: will machines replace humans? That is already happening. The critical question is: what happens to a society when economic participation stops being the default condition for a large share of the population?
The “useless class”: economic irrelevance, not social worth
In this narrative, the central issue is not unemployment as a temporary economic variable. It is the possibility that human beings become what Yuval Noah Harari provocatively described as a “class of irrelevants” or “useless”—in the economic sense, not necessarily in the social sense.
This distinction matters. People may remain meaningful inside their families, friendships, and communities. The risk is different: they may lose economic utility within the way the economy is organised. Their jobs disappear, not because they are incapable, but because work is executed more efficiently, more cheaply, and more scalably by machines—digital, virtual, physical, or hybrid.
The deeper implication is that power shifts. If a growing share of value is produced with minimal human labour, large segments of the population lose bargaining capacity. They become less “needed” by the productive system. That does not only change employment statistics—it changes the foundations of the social contract.
And once the social contract begins to fracture, the system starts producing secondary effects: polarisation, scapegoating, populist cycles, fragmentation of trust, and the kind of political volatility that is difficult to reverse.
When value stops flowing through work
The critical moment in this narrative emerges when the phenomenon becomes structural. Market economies historically distribute a portion of wealth through work and employment. Work is not only production; it is also one of society’s main mechanisms of value transmission.
If employment shrinks significantly, those transmission belts start to fail. Three mechanisms are particularly central—and each one becomes unstable under sustained job contraction.
1) Wages: the primary distribution channel weakens
Wages are still the main way most people access the value created by an economy. If jobs disappear, wages stop operating as a broad redistribution mechanism.
This is one reason why the future of work cannot be analysed only in terms of productivity. Productivity may rise dramatically, but if wages no longer connect people to value creation, productivity can coexist with exclusion. In fact, in this narrative, the paradox becomes acute: society may become richer while more people become economically unnecessary.
2) Taxation and social security: the fiscal architecture breaks
Modern welfare states are deeply linked—directly or indirectly—to labour income and employment-linked contributions. In an economy with less human employment, tax and social security systems would need radical redesign to preserve stability and legitimacy.
This becomes particularly visible in pension systems, especially pay-as-you-go models that assume stable employment and continuous contribution across decades. If employment contracts structurally, maintaining such systems becomes increasingly unsustainable without major reform.
The question then becomes: What do we tax, and how do we finance protection?
Labour may no longer be the main base. Capital, high-automation profits, rents, and new forms of value creation would have to carry a much heavier role. And that implies political conflict, because it forces societies to renegotiate who carries the burden of social continuity.
3) Direct access to capital income: participation becomes critical
If employment no longer distributes value, societies face a second-order shift: people may need to access wealth directly through ownership mechanisms—equities, funds, profit-sharing vehicles, collective ownership models, or other instruments.
This is often described as the “democratisation of investing.” And it is true that access has broadened: platforms and regulatory changes have made investment tools available to almost anyone—something far less common 10–20 years ago.
But here a paradox appears. Access is not the same as participation. Without wages, many people lack the starting capital to invest. So a world where wealth increasingly flows through ownership can become a world where those without employment become doubly excluded: excluded from wages and excluded from capital returns.
This is one reason why “financial literacy” becomes more than personal optimisation. It becomes part of citizenship in a system where value distribution may increasingly depend on the ability to participate in capital flows.
The need for alternative value-transfer models
In this context, alternative models of economic and social organisation move from the margins to the centre. Proposals such as Universal Basic Income (UBI) become strategically relevant—not as ideology, but as a potential stabilising architecture to keep societies functional when wage-based distribution weakens.
But the most important point is that UBI is difficult, because it forces a redesign of the economic constitution: fiscal logic, legitimacy, governance, and political coalition-building. It is not a policy tweak. It is a structural transformation.
And even if the economic layer is solved, this narrative still produces a deeper disruption: meaning.
The collapse of work as the axis of identity
Work is not only a contract. It is an identity system. Many people build purpose and social recognition around employment: “I am a doctor,” “I am an economist,” “I am a manager,” “I am a musician,” “I am a police officer,” “I am a firefighter.”
If a large share of the population no longer has a job as we currently define it, societies face a profound change in how people define themselves, how they experience dignity, and how they construct meaning through the life course.
This leads to a third and deeply human question:
If, in the future, my basic material conditions are secured without needing to work, what will I do to feel fulfilled?
What replaces work as the source of purpose, belonging, recognition, and meaning?
This is one of the largest open questions in any serious reflection on a world where employment is no longer the central axis of social and economic life.
Because in this narrative, even “solving income” does not automatically solve existence.
Narrative 2 — Leap of Consciousness
Creative destruction and the emergence of a new layer of work
The second narrative accepts the continuation of technological acceleration, but interprets its effects through a familiar historical pattern: major industrial and technological revolutions destroy jobs—but they also create new ones. In this view, job extinction is real, but it is accompanied by job creation at scale.
However, this narrative is not a comforting cliché. For it to be plausible, it must be analysed at more than one level. The optimism is not automatic. It depends on timing, scale, and the nature of the new work that emerges.
Here, a crucial distinction from Narrative 1 emerges: this is not a world where machines replace humans completely. It is a world where machines take over large portions of cognitive execution—and humans are forced to move up the value chain into domains where human uniqueness still matters.
Level one: timing is everything
Even if new jobs emerge in sufficient quantity in the long run, the decisive question remains: when will they be created, and at what pace?
We can imagine that in the long run the system reaches a new equilibrium—an almost homeostatic state in economic terms. But before that equilibrium, it is entirely plausible that unemployment spikes to socially difficult levels. This means the transition itself becomes the core risk.
In other words: the problem is not only destination; it is the bridge.
This is why training and reskilling are not optional. They are the infrastructure of transition. And if countries underinvest in continuous training—if organisations treat it as secondary—the “bridge” becomes narrower, and the transition becomes harsher.
In this scenario, continuous learning becomes the new oxygen. Not as a slogan, but as a survival condition for employability and relevance.
Level two: how new work could emerge at scale
The deeper question is: how can new jobs emerge at the scale needed to replace those that disappear?
Here we can develop two complementary lines of reasoning. The first is counterintuitive: the new technology platforms may lead not to less employment, but—paradoxically—to more employment.
The “layer” logic: infrastructure creates new economic surfaces
Once language, image, video and real world models combined with the new robotics platforms become generic and ubiquitous, a new technological layer forms—one that automates tasks previously performed by humans. But the key idea is what happens above that layer: new tasks and new activities become possible that are not only absent today—they are difficult to imagine as economically relevant.
A close analogy exists in the rise of social media. When social platforms emerged as infrastructure, they opened spaces for services and occupations that previously had no economic form. Fifteen years ago, it was not intuitive to think of “influencer,” “podcaster,” “streamer,” or “gamer” as significant economic roles. Today they are consolidated activities, sometimes powerful sectors with cultural and economic weight.
The point is not that the future of work becomes entertainment. The point is the mechanism: infrastructure changes what can be monetised.
And it changes it in unpredictable ways. What becomes valuable is not merely what is efficient, but what is scarce. And scarcity shifts.
Natural language as a new interface for production
Apply the same logic to the new infrastructure created by neural networks and generative models. We can anticipate a future where building digital products becomes accessible through natural language. That would disrupt an entire domain historically built on technical scarcity: software development.
The classic example is programmers. For decades, programmers built applications, websites, and digital products by writing code. If technical development becomes widely accessible through natural language interfaces, the barrier to entry collapses. People stop “writing code” and start “programming” by describing outcomes, iterating with systems, and orchestrating tools.
This does not eliminate professionals. It changes what professionalism means—and it radically expands the population of potential builders.
It also redefines what becomes “high value.” When execution becomes cheap, value shifts toward: conceptual clarity, domain depth, taste, judgement, ethics, and the ability to define problems worth solving.
The shift in critical skills: from execution to human distinctiveness
If execution becomes automated, the centre of gravity of critical skills moves. Instead of valuing technical execution alone, economies may increasingly value capabilities that remain more singular to humans and less substitutable by cognitive automation:
creativity and imagination
critical thinking and epistemic discipline
the ability to create something genuinely new (not merely combine templates)
empathy and perspective-taking
a form of self-awareness and awareness of others that supports trust, care, and meaning
In my view, one of the most underappreciated shifts is this: as answers become abundant, questions become scarce. AI can produce responses in seconds. But it cannot fully replace the human capacity to ask the right questions, grounded in lived experience, contextual awareness, relational intelligence, and an embodied sense of what matters.
This is why I call it a “Leap of Consciousness.” It points to a future where economic premium shifts toward the human capacity to create meaning and relational value—not just functional output.
The dialectic between synthetic intelligence and human consciousness
A second, closely related line of reasoning concerns the relationship—not necessarily confrontational, but dialectical—between synthetic intelligence and human consciousness.
As long as there is a domain that remains exclusively human, or at least not fully accessible to technology, new intrinsically human activities may flourish above the automation layer. In parallel, certain roles may gain value not because they require the new infrastructure, but because they respond to human needs intensified by the digital world: care, wellbeing, belonging, identity work, and trust-building.
This is also why emotional resilience and mental health become central to employability. In accelerated systems, the stress does not come only from work intensity. It comes from constant reinvention, constant identity recalibration, and constant learning under pressure.
In this narrative, psychological capability becomes economic capability.
The central difficulty: we cannot foresee the new jobs
The obvious challenge is that we cannot clearly anticipate these future occupations. It is extremely difficult to predict which functions will become economically relevant in a world of intensive digital and physical automation.
Yet there are signals suggesting that the “space of human consciousness” is progressively monetised in multiple dimensions. Returning to the influencer example, many of these activities convert human dynamics—hedonism, recognition-seeking, interaction needs, aesthetic and symbolic aspirations, lifestyle signalling, and brand relationships—into economic value at scale.
This raises a critical tension: the creation of new jobs may not be guided by traditional notions of productivity or social utility, but by attention, emotion, identity, and experience economies.
And that may be a feature, not a bug—because those domains represent aspects of humanity that remain difficult to mechanise.
The risk: creation may not match destruction in the short-to-medium term
Even if new jobs emerge, it is equally plausible that the rhythm and scale of creation will not be sufficient—at least in the short and medium term—to compensate for automation-driven job destruction.
And the jobs most exposed are already visible:
customer support and service functions (e.g., call centres)
administrative and back-office operations
routine information processing and standardised reporting
entry-level cognitive “white-collar” roles
parts of professional services based on repetitive, standardisable tasks (including accounting)
Even software roles—once seen as scarce and protected—are being reconfigured. The transition will therefore be uneven and psychologically demanding: it disrupts not only jobs, but professional identities and social status ladders.
Which brings us to the third narrative: the one many people quietly hope for.
Narrative 3 — Managed Deceleration
When regulation, geopolitics, or shock slows adoption
The third narrative corresponds to a widespread intuition: perhaps the technological transformation around AI, robotics, and related platforms will not continue exponentially, but will become more incremental and slow. For this to happen, something must change in the external environment.
In practice, this narrative is not “technology stops.” It is: technology is slowed, governed, contained, and shaped.
The brake: regulation and policy at global scale
A meaningful slowdown likely requires intensified regulatory pressure and public policy intervention—possibly at a global level. This is difficult, because the world is currently operating under an effective arms-race logic between the United States and China. In such a context, investment incentives are powerful, and coordination is structurally hard.
This is why many argue that a true brake may only come after a serious shock: a disaster or extreme event that forces coordinated governance—such as catastrophic misuse of AI, for example in biological risks or other malicious applications.
A “positive” version of this scenario would require an event severe enough to alter investment and adoption behaviour, but not so catastrophic as to become civilisation-threatening. A warning strong enough to force the world to sit at the same table—without destroying the table itself.
If such a turning point emerged, it would likely push two agendas to the centre:
Containment: controlling the evolution and diffusion of systems
Alignment: ensuring these systems are aligned with human values and do not make decisions against human interests
The advantage of deceleration: time
If this narrative materialises, society gains its rarest strategic asset: time.
Time to adapt education systems and curricula.
Time to test redistribution and transition mechanisms.
Time to redesign labour law and social protection for more fluid careers.
Time for organisations to transition operating models with less trauma.
Job substitution could be mitigated, allowing more gradual management of skill obsolescence.
But time is only an advantage if it is used deliberately. Otherwise, deceleration becomes procrastination.
The reality: even incremental change is still significant
Even in a slower evolution path, transformation remains substantial. Organisations, leaders, individuals, and public policies still need to prepare—because the direction of change is preserved even if the speed is moderated.
The key difference is that deceleration increases the chance that institutions can keep up—if they use the time deliberately rather than waste it.
In this narrative, the strategic failure is not being disrupted too fast. It is being given time—and failing to reform anyway.
Conclusion: three narratives, one decisive variable
These three narratives describe different futures, but they converge around one decisive variable: adaptation capacity.
How quickly can societies redesign value-transfer mechanisms if wages weaken?
How fast can education and reskilling systems operate at scale?
Can labour market protections follow people across fluid careers?
Can individuals rebuild identity and meaning beyond stable employment?
Can governance keep pace with acceleration without becoming purely reactive?
The future of work is not only a technological question. It is a political economy question—and, ultimately, a human meaning question.
And that is why “optimism” or “pessimism” is the wrong starting point.
The right starting point is preparedness.
Key takeaways
Net job numbers can be misleading. Timing, distribution, and transition capacity matter more than totals.
Narrative 1 (Revolution of the Irrelevant) is triggered when automation outpaces re-skilling and social adaptation, weakening wages, fiscal systems, and identity structures.
Narrative 2 (Leap of Consciousness) is credible only if job creation emerges at scale and fast enough—and if societies shift skill formation toward human distinctiveness: questions, meaning-making, empathy, judgement.
Narrative 3 (Managed Deceleration) is plausible only if governance or geopolitics meaningfully slows adoption—often after a shock—and if societies use “time” as a reform asset.
Across all narratives, adaptation capacity is the master variable—for individuals, organisations, and governments.
Questions to readers
Which of the three narratives feels most plausible in your sector—and why?
What early signals would convince you that Narrative 1 is unfolding (structural rupture), rather than Narrative 2 (creative reconstruction)?
If “answers become abundant,” what practices help you cultivate better questions?
What would a credible new value-transfer model look like in your country if wages weaken structurally?
If work becomes less central to identity, where do you think purpose, belonging, and recognition will migrate?
If you reply, tell me which narrative you are preparing for—and what you are changing now because of it.
This essay synthesises insights from multiple research and advisory projects conducted over the last two years by IF Insight & Foresight and The Long Game, focused on the Future of Work—spanning horizon scanning, interviews, scenario development, and strategic implications for organisations and public institutions.
Short bibliographic references
World Economic Forum. (2025). The Future of Jobs Report 2025.
International Labour Organization. (2023). Generative AI and Jobs: A global analysis of potential effects on job quantity and quality.
International Labour Organization. (2025). Generative AI and Jobs: A refined global index of occupational exposure (2025 update).
OECD. (2023). OECD Employment Outlook 2023 (incl. chapter on AI and the labour market).
Harari, Y. N. (2017). Homo Deus: A Brief History of Tomorrow. (discussion of a potential “useless class”).
Brynjolfsson, E., & McAfee, A. (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies.
I work in the healthcare data and insights industry, and I am preparing for narrative #2. Healthcare data and insights are both complex and complicated, which makes the industry ideally suited to people who have very good critical thinking and are able to consider 2nd, 3rd and 4th order problems as part of their work. AI (in my opinion, especially agentic AI) will be excellent at bringing efficiency to some of the tasks that the work requires, but the processes of bringing it all together in meaningful ways will require high-quality humans for some time to come.
Happy to talk about this further (I have followed you and requested a connection).
Thanks for the excellent work.