Artificial intelligence is being blamed for making experienced workers obsolete. The rather more uncomfortable truth is that rapidly changing industries have always favoured people who can learn quickly, adapt freely and abandon ideas that no longer work. AI has not created that reality. It has merely made it much harder to ignore.
There is an increasingly common observation that older people, particularly those working in technology and other rapidly changing fields, are finding it harder to secure senior roles. Artificial intelligence is the current ‘big thing’ so is often presented as the cause. Employers want people who understand AI, the argument goes, and experienced professionals are being pushed aside by a younger generation more comfortable with the technology.
I am not convinced.
Admittedly, my first hypothesis was exactly what I said above – that older people have experience that is largely outdated in this new age (and I am amongst the people I so malign). On think deeper that was far to blunt an analysis. It’s not because older workers are necessarily keeping pace with AI. Nor is it because age discrimination is imaginary, that’s obvious, and protected in law in the UK’s under the Equality Act 2010; it applies to job applicants as well as existing and former workers. Nevertheless. a House of Commons inquiry into older people and employment previously found that witnesses regarded age bias and discrimination, particularly during recruitment, as a significant barrier to older people remaining in work.
The young are not having an easy time either
It is worth acknowledging that youth is hardly a golden ticket in today’s labour market. While much of the discussion around AI focuses on whether experienced professionals can remain relevant, a different challenge is unfolding at the other end of the age spectrum. In early 2026, more than one million people aged 16 to 24 in the UK were not in education, employment or training, the first time that figure had exceeded one million since 2013. Youth unemployment also rose significantly, with around 739,000 young people aged 16 to 24 recorded as unemployed. These figures sit rather awkwardly alongside the suggestion that young people are displacing older workers.
Part of the answer may be that organisations increasingly want contradictory things. They want adaptability, but they also want experience. They want fresh thinking, but they also want proven results. They want potential, but they prefer somebody else to have paid for its development.
This creates an interesting paradox. Older workers are often told their experience is becoming obsolete. Younger workers are frequently told they lack experience altogether. One group is criticised for knowing too much about the old world. The other is criticised for not knowing enough about the current one.
Older employees may be passed over for knowing too much about the old world.
Younger ones for not knowing enough about the current one.
Perhaps the deeper issue is that disruption has broken the traditional career ladder. For decades, the assumption was simple. Young people would enter organisations, learn their craft, gain experience and gradually progress into more senior roles. Experience accumulated, responsibilities increased, and careers largely moved in one direction. Today, both ends of that journey appear less certain.
Many younger people struggle to secure stable employment, while older people discover that some of the skills and experience they spent years acquiring are no longer valued in the same way. The ladder still exists, but some of the lower rungs are missing and some of the upper ones have been quietly removed.
AI did not create this situation. Globalisation, automation, outsourcing, changing business models and the accelerating pace of technological change were already reshaping careers long before generative AI appeared. What AI is doing, with its remarkable pace and impact, is accelerating trends that were already underway from the start of the Industrial Revolution, and especially in the Communication Age; exposing weaknesses in the assumptions upon which many career paths were built. Viewed through that lens, the challenge is not really about age at all. It is about how organisations identify, develop and reward capability in a world where both knowledge and technology have increasingly short shelf lives.
My difficulty is with the suggestion that AI created the problem. In technical roles, the uncomfortable relationship between age, experience and continuing relevance existed long before ChatGPT appeared. AI has accelerated the phenomenon and made it more visible, but it did not invent it.
Experience alone is no longer sufficient.
Equally, potential alone is not enough.
However, it’s only certain types of skill that age poorly; often it’s the ‘technical’ ones. The people who appear most likely to thrive are those who can combine curiosity with judgement, learning with application, and adaptability with the ability to deliver real-world outcomes. These are skills that can accumulate over time.
Experience has always been a depreciating asset
We talk about experience as though it were a steadily accumulating store of value. Every year adds another layer and, provided we continue turning up, we become progressively more useful. That has been true, in stable environments. Where tomorrow’s problems resemble yesterday’s, experience allows us to recognise patterns, avoid familiar mistakes and make good decisions without laboriously rediscovering everything from first principles.
However, the value of experience is conditional upon the future retaining some resemblance to the past.
Experience is enormously valuable when the future resembles the past. When the world changes, some experience becomes baggage.
This has been true throughout the history of technology. Expertise in a product, platform or working method can move from being scarce and valuable to commonplace, automated or irrelevant. The practitioner may be no less intelligent, diligent or capable than before, but the market value of what they know has changed around them. They may not realise it, which is problematic. There is no moral judgement; no one becomes less deserving because a technology changes. It is simply the way that markets value specific capabilities. Experience does not have an intrinsic market value. It has contextual value because it helps an organisation address the problems it faces now and anticipates facing next.
Artificial intelligence makes this depreciation unusually obvious. The tools, models, products, terminology and accepted practices are developing so quickly that even quite recently acquired knowledge can become dated. A course written around a particular AI interface, prompt technique or model capability may remain interesting, but cease to be operationally useful surprisingly quickly. Even meta-knowledge about AI – how to leverage it, good practices, adoption expertise – is aging quicker than we expect. Event to the point where the need goes away, replaced by another iteration or innovation of the platform technology.
Can anyone really be experienced in modern AI?
I recently discussed this with a colleague; whether anyone could meaningfully claim to possess substantial experience of the current generation of AI.
There are, of course, people with long and distinguished careers in machine learning, natural language processing, data science and related disciplines. Their expertise is genuine and probably has reasonable longevity. There are also people with considerable practical experience of implementing automation, managing information, developing software and leading complex organisational change. None of that should be casually dismissed. Nevertheless, the widespread business use of generative AI is very young. The practical environment in which most organisations are now operating did not exist in its current form five years ago, or even at the start of this year. Important capabilities, constraints and working practices are changing on a monthly cadence. Anyone who claims settled expertise in the whole field should probably be treated with at least a little suspicion.
The UK Government’s 2026 rapid evidence review of AI skills for life and work describes an increasingly urgent need for AI skills, but also stresses the importance of broader digital literacy and notes that attainment is influenced by age, education, income, region and other demographic factors. It does not describe AI capability as one neat skill which a person either possesses or lacks. It is an evolving collection of technical, applied, social and critical capabilities. That changes what organisations should look for. In an unsettled field, the ability to recite yesterday’s best practice may be less useful than the ability to work out why it no longer applies.
The best young people have a genuine advantage
At this point, discussions about age at work often become unnecessarily polite. We reassure ourselves that every generation brings equal but different strengths, and that age is simply a number. It is a pleasant sentiment, but not an entirely convincing account of how competitive employment markets operate.
The best young people may well be more useful to many organisations than ordinary experienced older people. They can be astonishingly quick to understand new ideas, less encumbered by assumptions about how work ought to be done, more willing to experiment and more comfortable operating without an established map. They are also generally cheaper to employ.
That last point should not be ignored. An experienced professional may bring decades of accumulated knowledge, but will commonly expect a salary, and the seniority, that reflects it. If much of that knowledge is no longer central to the organisation’s needs, the economic proposition becomes difficult. An exceptional younger person who can learn the role quickly, challenge its assumptions and grow with the organisation may simply represent better value.
It goes without saying that this is a generalisation; not all young people are intellectually agile, ambitious or even especially interested in learning. Not all older people become rigid, complacent or technically irrelevant. The comparison is not really between all younger people and all older people. It is between particular individuals with different combinations of capability, knowledge, curiosity, judgement and cost. Even so, applying the necessary caveats should not prevent us from acknowledging an uncomfortable probability. Physical youth arrives automatically. Intellectual youth has to be maintained deliberately.
Younger does not simply mean smarter
Describing younger people as “smarter” is, however, too crude. Intelligence does not rise to a single peak and then uniformly decline. Different capabilities follow different trajectories.
The distinction between fluid and crystallised intelligence is helpful here. Fluid intelligence includes the ability to process unfamiliar information, reason abstractly and solve novel problems. Crystallised intelligence reflects accumulated knowledge, language, strategies and understanding. The former tends to favour speed and novelty; the latter is built through learning and experience.
A summary from the Stanford Center on Longevity describes human capability not as a single summit but as a mountain range, with different attributes peaking at different ages. Its review reports that raw processing speed tends to peak earlier, while capabilities such as emotional intelligence, conscientiousness, emotional stability and moral reasoning may develop or remain strong later in life.
The important point is not that one age group is definitively more intelligent than another. It is that they are likely to bring different portfolios of cognitive strengths. A talented younger professional may understand a new tool more rapidly. An experienced colleague may be better able to recognise why the tool is being introduced, which apparent problem is merely a symptom, how people will respond, where governance will fail and which supposedly revolutionary idea has already failed twice under different names.
One produces highly visible capability. The other may prevent an expensive and largely invisible mistake.
The best experience is experience of disruption
This is where useful experience distinguishes itself from the mere passage of time. Knowing the detailed behaviour of an obsolete product may have little remaining value. Having lived through several waves of technological disruption can be extraordinarily valuable, provided the individual has been paying attention. That’s essential meta-knowledge.
Those of us who have worked through personal computing, client-server systems, the web, mobile devices, cloud services, social platforms, software as a service, low-code tools and now generative AI have seen certain patterns recur. We have watched new technologies arrive surrounded by implausible claims. We have seen organisations mistake buying technology for adopting it. We have seen expensive transformation programmes fail because nobody addressed behaviour, information quality, governance, incentives or trust.
None of that experience tells us exactly what AI will do next. It does, however, help us ask better questions. The value lies not in having survived a long time (though that is something to be grateful for), but in having repeatedly adapted. Someone who has spent twenty years reconsidering assumptions, changing tools and learning new disciplines has not merely acquired experience, they have practised reinvention.
Twenty years of continuous learning is valuable. One year of experience repeated twenty times is not.
Or, as Einstein observed, “We cannot solve our problems with the same thinking we used when we created them.” In a world reshaped by AI, the challenge facing both young and experienced workers is not a lack of intelligence or effort, but the willingness to rethink how expertise is acquired, maintained and applied.
AI problems are rarely just AI problems
There is a further reason not to discard experience too readily. Many of the hardest problems associated with AI are not principally technical. They concern the quality and structure of organisational information, the design of work, the reliability of decisions, the ownership of risk, the behaviour of leaders, the willingness of employees to change and the gap between what a system can demonstrate and what it can deliver consistently in a real organisation.
AI can generate an impressive answer in seconds. Deciding whether the answer is correct, useful, lawful, proportionate and appropriate may require substantial contextual understanding. It may require somebody who remembers why a particular control exists, understands how an apparently local change affects the wider system, and knows which stakeholder will quietly bring the whole endeavour to a halt if ignored.
Deciding whether the answer is correct, useful, lawful, proportionate and appropriate may require substantial contextual understanding.
Experience is especially valuable here, but only when it remains connected to curiosity. Experience without curiosity becomes dogma. Curiosity without experience can become expensive enthusiasm.
The real divide is between learners and non-learners
It is tempting to rescue the argument by declaring that age is irrelevant and mindset is everything. Age is not irrelevant; generally, some kinds of cognitive processing change as we grow older. Meanwhile, salary expectations rise, personal circumstances tend to reduce appetite for risk. Deep expertise can make it psychologically harder to return to being a beginner. The longer a particular way of working has rewarded us, the harder it can be to accept that it no longer deserves to.
A more useful distinction is between people who treat learning as something they completed and people who regard it as part of their professional identity. There are thirty-year-olds whose expertise has already hardened into certainty. There are sixty-year-olds who are still exploring, questioning and rebuilding their understanding. The former may be chronologically young but intellectually old. The latter have done something more difficult than being born recently; they remained actively curious.
The skills most likely to retain their value are not skills associated with a particular AI product, they are the capabilities that allow somebody to operate when the product, problem and accepted practice all change:
- learning quickly without pretending to know more than they do;
- thinking critically about confident but weakly supported claims;
- separating technological possibility from organisational value;
- recognising systems, consequences and dependencies;
- abandoning ideas that have ceased to work;
- communicating uncertain and complex issues clearly;
- combining informed scepticism with a willingness to experiment.
These are certainly not capabilities aligned with youth or age; they do arise through practise. Younger people, however, enjoy a natural advantage because they have less to unlearn and less status invested in the previous answer.
What should organisations do differently?
Organisations should resist two equally lazy forms of recruitment:
The first assumes that years of experience are an adequate proxy for likely performance. Asking for ten years’ experience in a field whose current form barely existed five years ago is not rigorous recruitment. It is numerically decorated nonsense; something I have also seen in RFPs (request for proposal).
The second assumes that youth is an adequate proxy for adaptability. Hiring somebody because they appear digitally native risks replacing one stereotype with another. Familiarity with fashionable tools does not guarantee judgement, persistence, critical thinking or the ability to achieve change through other people.
Recruitment for disrupted roles should examine how candidates learn. Ask what they have changed their minds about. Explore how they respond when their expertise stops working. Look for evidence that they can transfer understanding from one domain to another. Test whether they can distinguish a convincing demonstration from a sustainable operating model. Ask them to explain what they do not yet understand.
Organisations should also stop treating learning as a periodic remedial activity for people who have fallen behind. The UK Government’s AI Skills for Life and Work summary report draws on an evidence review, public and employer surveys, vacancy analysis, expert study and stakeholder workshops. Its scope itself demonstrates the breadth of the challenge. AI readiness is not achieved by sending a few specialists on a technical course. It involves developing capability across the workforce and understanding how requirements will continue to change. As such, learning must become part of work rather than an occasional interruption to it. Experimentation needs permission, time and appropriate guardrails. Experienced employees need opportunities to become beginners without feeling that doing so diminishes their status. Younger employees need access to the context and organisational memory that prevent speed from becoming recklessness.
The future belongs to the intellectually young
So, are the best young people likely to be more useful to organisations than ordinary experienced older people?
Probably.
The best of any group will commonly outperform the ordinary members of another, and talented younger people may combine intelligence, learning speed, adaptability, energy and lower employment cost in a form that is extremely attractive to organisations navigating disruption. But that’s not the end of the argument. The strongest older professionals are not competing by trying to be twenty-five again. They are combining continued learning with pattern recognition, context, judgement and experience of previous disruption. Likewise, the strongest younger professionals are not valuable merely because they learn a new interface quickly. They are valuable because they can use that speed to develop understanding rather than simply accumulate technique.
AI may well be reducing the value of such experience as has become detached from learning, regardless of age. The future does not belong automatically to the young, nor does seniority guarantee a place in it.
It belongs to people who can remain intellectually young as they age.
That is harder than simply being young in the first place. It is also a choice we have to keep making.
Sources and further reading
- Acas: Age discrimination and what the law says
- House of Commons Women and Equalities Committee: Older people and employment
- Stanford Center on Longevity: Fluid and crystallised intelligence
- UK Government: AI Skills for Life and Work rapid evidence review
- UK Government: AI Skills for Life and Work summary report
- Pearn Kandola: Age Discrimination at Work report
- Young people not in education, employment or training (NEET), UK: May 2026
Novia Works Ltd 