The foundation of modern employment, study hard, secure a job, build a stable life, is eroding fast. Technological progress, particularly in artificial intelligence (AI) and automation, is breaking down the old economic model that tied individual effort to secure long term work. From logistics to legal services, tasks considered immune to disruption are now being replaced by algorithms at a fraction of the cost and time.
This shift is triggering a structural imbalance across the globe. Highly qualified professionals, graduates, Master’s and Doctorate holders are increasingly found in gig-based roles far beneath their skillsets. Meanwhile, traditional full-time employment is giving way to precarious, mobile phone app-based work where benefits, career progression and long-term stability do not exist.
At the centre of this transformation is clear economic reasoning. Hiring people has become expensive, machines are not. Governments have unwittingly incentivised automation by taxing labour and subsidising capital investment. Businesses, under pressure to reduce costs and boost efficiency, are responding rationally by reducing headcount and investing in AI.
The implications of that response? Who remains employable in a market where qualifications are devalued? Is academic attainment still relevant in determining economic worth? And as machines displace both manual and cognitive labour, what kinds of roles, if any, can be considered future proof?
The state of employment in the West: oversupply, underemployment and the gig economy
In the UK, US and across Europe, labour markets are experiencing a dual shock; an oversupply of educated workers and a shrinking pool of stable employment opportunities. Data from the US Bureau of Labor Statistics and supporting analyses reveal that over 41 million Americans were not earning a living wage even before the pandemic, a number which has only grown. A substantial share of these hold advanced degrees but are now employed in the gig economy, and not by choice.
The picture is similar in the UK, where Office for National Statistics (ONS) figures show a consistent rise in underemployment among university graduates. Highly educated individuals are increasingly found in roles that require no more than a driving licence or a smartphone, like food delivery, ride-hailing and warehouse stacking.
This structural shift is not limited to the lowest rungs of the economy. Mid-tier professional roles in law, accounting and marketing are being gradually stripped of headcount as companies adopt AI tools that perform tasks faster, cheaper and around the clock.
The erosion of secure, salaried positions and the normalisation of gig-based employment mark a profound departure from the 20th-century labour contract. For Western economies, the long-term risk is a growing class of educated but economically insecure workers that are overqualified and increasingly disillusioned.
Entrepreneurship and freelancing: a survival strategy or a mirage?
With stable employment increasingly out of reach, millions are turning to freelancing and self-employment, more often as a last resort than a chosen path. In the US alone, more than 70 million people now identify as freelancers, according to data from Upwork and WorkPajama. In the UK, over 4.4 million are self-employed, yet a significant share remains underutilised and underpaid.
Beneath the headlines celebrating entrepreneurial spirit lies a more sobering reality. Many who launch online businesses or freelance ventures are doing so out of economic pressure, not passion. They are attempting to monetise hobbies, skills or time, often while juggling part-time jobs or gig work. But in saturated digital marketplaces – where the supply of sellers dwarfs the consumer demand – few generate sustainable income. Social media is awash with micro-enterprises offering the same digital services or handmade goods, while consumer purchasing power remains limited.
Freelancers face an environment of deep insecurity. Without access to employer-sponsored healthcare, pensions or legal protections, many operate without a safety net. Income is inconsistent. Workloads are unpredictable. Burnout is common. For every story of an online business success, there are thousands quietly closing shop.
What emerges is a freelance economy that promises independence but often delivers little. For policymakers and business leaders alike, the challenge is to consider what is entrepreneurial spirit and what is a labour market in decline, where self-employment is not a sign of growth, but of structural failure.
Cutting costs with code: why companies prefer machines over people
The move toward automation is not driven by ideology for its own sake. It has become a rational financial business decision. For many employers, AI and automated systems are simply more efficient and less costly than human labour.
Human workers come with overheads like salaries, taxes, pension contributions, healthcare, annual leave and regulatory obligations. For companies already operating on thin margins or under investor pressure to reduce costs, these cumulative expenses turn human labour into a liability. By contrast, machines require no sick days, do not unionise, and can operate continuously without supervision or complaint.
Governments have compounded this imbalance. Capital investments in automation are now rewarded with tax breaks and depreciation benefits, while labour is heavily taxed and regulated. In effect, the policy frameworks themselves across Western economies have made employing people more expensive than building systems to replace them.
The result is a quiet but consistent shift. Legal firms are turning to contract analysis tools. Accounting is increasingly handled by cloud-based AI. Content creation, market research, logistics co-ordination, and customer service fields previously protected by experience and complexity, are being absorbed into digital workflows. In some sectors, this transition is immediate. In others, it is incremental, under the guise of board level strategies that see existing staff forced to take on more responsibility across multiple functions for the same salary.
Moreover, cross-functional staffing is now being seen more often. At airports, a single employee may check baggage, guide passengers and manage gate boarding. In US restaurants, chefs are being made to deliver food to tables to eliminate waiting staff.
These changes are framed as operational efficiency but in reality, they are cost prevention methods designed to take a step back from full automation while still reducing headcount.
Technology may not replace every job at once, but the economic logic is already rewriting the rules of employment. For businesses, automation is not a choice, it’s a necessity.
A generational divide: who embraces AI and who resists it?
A stark generational split is emerging in how artificial intelligence is understood and adopted in the workplace. Sam Altman, CEO of OpenAI, has drawn attention to the way age correlates with AI usage. The implications for business are far from trivial.
Those aged between 0 and 25 are already integrating AI into their daily lives with fluency and creativity. For them, tools like ChatGPT or Claude are not novelties but essential utilities, used to plan schedules, solve academic or technical problems, write code, delegate tasks or even draft communications. This cohort is not learning to use AI, they are growing up and are at one with it.
By contrast, the 25–35 demographic engages with AI in a more casual, transactional way. They tend to use it to ask life questions, resolve minor dilemmas or conduct general internet searches, often as a replacement for Google. While familiar with the technology, this group is not as deeply embedded in its ecosystem.
It is the over-50s, however, who represent the widest adoption gap. In leadership positions, they often remain sceptical, viewing AI as overhyped, unreliable or a threat to business as usual. As decision-makers, this reluctance to explore or invest in AI will leave organisations vulnerable to stagnation.
The tsunami of highly intelligent and qualified under 35s that have embraced AI and have embedded it into their workplace are the future. The risks are clear. Companies led by AI-literate, adaptive teams are better positioned to innovate and capture efficiencies. Those led by executives resistant to technological change may find themselves outpaced, internally by younger staff or worse, externally by competitors. As AI continues to evolve, so too will the divide between those who see it as a tool and those who treat it as a threat.
When GDP no longer needs people: decoupling labour from economic growth
One of the most profound and least discussed developments of the AI era is the quiet uncoupling of economic growth from human employment. In traditional industrial models, rising GDP went hand in hand with job creation. More production meant more workers and higher wages, followed by increased consumption. But in the emerging post-labour economy, this link is breaking. Nations can now increase output not by employing more people, but by implementing more code.
This shift challenges the very premise of the social contract, being that individuals willing to work will find opportunities to contribute and earn. As AI systems become more capable across sectors, from manufacturing and logistics, to law, education and content production, the demand for human labour is not just changing, but in many cases, disappearing. The right to work becomes meaningless if no one is hiring humans.
This is not a dystopian prediction. It is already happening. In May 2025, Microsoft laid off approximately 6,000 employees, about 3% of its global workforce. The cuts spanned various departments, including LinkedIn and Xbox, and were part of a strategic shift towards AI investments. In February 2025, Meta laid off around 3,600 employees, representing 5% of its workforce, targeting “low performers” as part of efficiency measures. In early 2025, Amazon planned to eliminate 14,000 managerial positions to streamline operations and reduce costs by up to US$3.6 billion annually. In November 2022, under Elon Musk’s leadership, Twitter laid off approximately 3,750 employees, about 50% of its workforce, as part of cost-cutting measures.
It is clear that companies are generating record profits while cutting staff. GDPs are rising even as labour force participation flatlines. The logic of automation allows businesses to scale without hiring, optimise without training, and compete globally without relying on large human workforces. In this model, humans are not central to growth, they are a cost to be either managed or removed.
Yet, this new economic culture conceals a deeper structural risk, which is demand. A consumer economy relies on people not just working, but spending. If the labour force is increasingly excluded from income-generating roles, cumulative demand may collapse. Who will buy even the AI-made goods and services if vast swathes of the population are underemployed or excluded entirely from the economy?
This paradox – machines generating output that humans can no longer afford – could mark the breaking point for the economic model as we know it. Without clear structural reforms, economies risk becoming efficient but low-employment systems, where wealth is for the few and inequality becomes the norm.
The road ahead: what does the future of work look like?
The traditional career arc, “learn a skill, apply it for decades, then retire”, is all but gone. In its place emerges a model shaped by constant change, “learn, do, re-learn, repeat”. The accelerating pace of automation and the rapid evolution of technology demand a workforce capable of constant adaptation, not static mastery.
University degrees, once golden tickets to employment, are increasingly outpaced by the speed of technological disruption. In many fields, the tools and techniques learned in a four-year programme are obsolete by the time of graduation. What now matters more is the ability to learn fast, apply and adapt, qualities not taught by academia but by a new mindset.
As machines continue to outperform humans in both manual and cognitive tasks, the remaining economic territory for human labour is shrinking. What remains defensible are interdisciplinary skills, emotional intelligence and authenticity, areas where machines struggle to replicate empathy or morals. Workers who can integrate knowledge across domains, communicate with clarity or build trust in human relationships will likely remain valuable, even as technical roles yield to automation.
But even with this, survival is not guaranteed. The emerging ‘post-labour’ economy foresees a world with fewer jobs overall, especially in the middle. The polarisation of work into high-skill, high-paying roles and low-skill, low-paying ones, will widen. Meanwhile, sectors once thought automation-proof, education, counselling and journalism, are already seeing AI inroads. What may be left for humans is not a professional class, but a performance one of trusted advisors or charismatic guides.
In this environment, resilience and reinvention are the golden tickets. The ability to collaborate, to think critically, to empathise and to tell compelling stories, these will matter more than certifications or years studied. The future of work will not be about competing with machines but about excelling at what they cannot do.
Conclusion: the question is not If, but when
The unravelling of the employment contract is no longer theoretical, it is underway. The labour market has not merely changed; it has begun to fragment. The assumptions that once held firm, that education guarantees opportunity, that full-time work ensures stability, that hard effort leads to reward, are no longer universally true.
For most people, the traditional idea of work is being rewritten in real time. Employment has shifted from a stable destination to a precarious condition, from a collective structure to an individual struggle. The language of job security and career progression is being replaced by that of gig tasks and side hustles. Yet, within this upheaval lies something very important. The future will not be built by resisting automation or dwelling on the past. It will be written by those who adapt quickly and lead.
Governments that rethink policy, companies that retrain rather than replace, and individuals who reinvent rather than retreat, will do more than survive. But for everyone else, the clock is ticking. The question is no longer if the world of work will change. It is how soon and who will be ready?
To download a copy of this article, please click here.
