Work, ownership, and the new architecture of economic security
By Po-Sung(Sinclair) Huang
For decades, people believed that working hard was enough.
In the AI era, that assumption is quietly breaking.
Not because work disappears, but because ownership matters more than ever.
In the AI era, relying on labour income alone is no longer a neutral choice. It is an active form of risk.
That sentence may sound harsh. But it captures a structural shift that many people can already feel, even if they do not yet have the language for it. They feel it in markets. In career anxiety. In the fear of being left behind by a technology wave they did not ask for, do not fully control, and may not directly benefit from.
This is why 2026 feels different.
Not because inequality suddenly began this year. It has been building for decades. But AI is making the underlying logic of the divide harder to ignore. The gains from technology are flowing not only through wages but through ownership, infrastructure, data, and capital.
That distinction matters. Recent work from the IMF suggests AI can raise productivity while also widening wealth inequality when returns to capital rise faster than benefits diffuse through wages. OECD research points in a similar direction: AI can improve productivity and some aspects of work, but adoption remains uneven, and the benefits depend heavily on skills, organisational readiness, and complementary systems.
This is why the central question of the AI era is no longer simply who can use the newest tools.
It is who owns the assets, systems, and bottlenecks that those tools depend on.
AI is not just software — it is a physical bottleneck in the economy
Too many discussions about AI remain trapped in the language of software alone: models, prompts, agents, copilots, apps. But the real AI economy is not made of software only. It is built on energy, semiconductors, memory, advanced packaging, interconnects, cooling, data centres, and the capital expenditure required to scale all of them.
In other words, AI is not just a software story. It is also a physical systems story.
That matters because physical systems create bottlenecks, and bottlenecks create pricing power.
The winners in an AI economy are not necessarily only those building the most impressive model. They may also include those closest to scarce infrastructure: compute, power, thermal management, bandwidth, packaging, and the industrial capacity to integrate these layers into real operating systems.
This is one of the most important shifts investors and non-investors alike need to understand. In earlier digital eras, it was easier to imagine value rushing toward software and away from the physical layer. In the AI era, that assumption is far less safe. The deeper AI penetrates the real economy, the more it depends on infrastructure that cannot be replicated instantly.
The most durable gains may not always belong to the loudest narratives.
They may belong to the owners of the quiet bottlenecks.
Work is no longer the anchor it once was
For much of the modern era, work did more than generate income. It organised life.
Education fed into jobs. Jobs fed into identity. Identity fed into belonging. Wages, over time, were supposed to support savings, family formation, and eventually modest asset accumulation.
That architecture is now under strain.
Jeremy Rifkin argued in The End of Work that technological progress could steadily erode the central place of human labour in economic life, raising the possibility of both liberation from drudgery and a more unequal, unstable order if the gains were concentrated in too few hands. He also pushed readers to think beyond market and state toward what he called a “post-market era,” with a larger role for civil society and the “third sector.”
Nearly three decades later, AI has made that argument feel newly immediate.
What makes AI different from many earlier digital waves is not merely its speed, but its reach. It touches white-collar work, analysis, writing, software development, design, decision support, and increasingly the coordination layer of organisations themselves. It not only automates routine tasks. It changes how intelligence, judgment, and execution are packaged and distributed across firms and systems.
Rifkin’s warning matters because it reminds us that the question is not only economic. It is civilizational. If work no longer guarantees dignity, belonging, and security in the way it once seemed to, then societies will eventually need to redefine value more broadly than paid employment alone.
But before societies solve that problem, individuals will feel the pressure first.
Jobs are no longer a guarantee
If Rifkin diagnosed a macro-level shift in the structure of labour, Taylor Pearson pushed the question into personal strategy.
In The End of Jobs, Pearson argues that traditional employment is becoming less reliable as the default path to economic security, and that entrepreneurship, ownership, and self-directed economic agency may in many cases be safer than dependence on a single employer.
That insight matters more in the AI era than ever.
The old middle-class bargain assumed that skills could be exchanged for a relatively stable institutional role. But when globalisation, software, AI, and falling barriers to entry reshape industries faster than organisations can promise stability, having a job is no longer equivalent to having security.
This does not mean everyone must become a founder in the conventional sense.
It does mean more people may need to think beyond jobs and more like owners: of assets, of systems, of reusable knowledge, of productive tools, and of income streams that are not tied entirely to hours worked.
The key shift is psychological before it is financial.
The goal is not to idolise entrepreneurship. It is to recognise that economic security is increasingly built less on position within an institution, and more on position within a system.
Labour earns. Assets compound.
None of this means labour no longer matters. Skills matter. Judgment matters. Domain expertise matters. The ability to work with AI, adapt to new tools, and operate inside increasingly complex systems will matter enormously.
But labour and assets are not the same thing.
Labour earns. Assets compound.
And in the AI era, the gap between the two may widen faster than most people expect.
That difference has always existed. What changes now is how visible and consequential it may become. If AI raises productivity while channelling a disproportionate share of gains toward capital owners, then even people with stable jobs may find that salary growth alone no longer keeps pace with the appreciation of productive assets.
This does not imply that everyone must become a trader or that every worker must transform into a portfolio strategist. It means something simpler and more uncomfortable: in a world increasingly driven by ownership, refusing to think about ownership is no longer a neutral choice.
For many people, the real question is not whether they can discover the next tenfold winner.
It is whether they can avoid standing entirely outside a system where more of the upside flows to those who own rather than only those who work.
What Poor Economics helps us see
This is where Poor Economics becomes essential, but only if read carefully.
One of the greatest mistakes in public discussions about inequality is the temptation to moralise too quickly. People are poor because they lack discipline. Because they think too short-term. Because they do not plan well. Because they make bad choices.
That is far too easy.
The work associated with Esther Duflo and Abhijit Banerjee, including J-PAL’s broader research tradition, is better understood as an effort to examine how incentives, constraints, incomplete information, and policy design shape real-world decisions under pressure. The 2019 Nobel Prize in Economic Sciences recognised Banerjee, Duflo, and Michael Kremer for their experimental approach to alleviating global poverty.
That difference is crucial.
One of the most important lessons associated with Poor Economics is that poverty is rarely just a failure of effort. More often, it is a condition of unstable incentives, thin margins, lack of buffers, and constant trade-offs. In such environments, long-term planning becomes harder not because people do not care about the future, but because the present is too unforgiving.
This matters enormously in the AI era.
If households already live close to the edge, they are less able to absorb volatility, less able to invest in retraining, less able to tolerate waiting for long-term returns, and less able to participate in the compounding of productive assets. They are more likely to optimise for immediate survival because immediate survival is the binding constraint.
That does not make them irrational.
It means the structure around them narrows the time horizon of rational action.
This is why inequality in the AI era should not be discussed only as a matter of talent or effort. It must also be understood as a matter of access: access to tools, access to training, access to buffers, access to ownership, and access to systems that allow knowledge to compound instead of merely being consumed.
What can ordinary people actually do?
This is the point where many readers will understandably ask: if I do not have venture capital, do not own a data centre, and cannot buy my way into the next wave of infrastructure, where do I begin?
The answer is not glamorous, but it is real.
First, broaden the meaning of “asset.” Ownership is not limited to securities. In the AI era, assets can also include reusable knowledge, scalable workflows, trusted relationships, content libraries, proprietary know-how, templates, code, niche expertise, and systems that continue to create value without requiring every unit of your time.
Second, treat broad-based ownership as more realistic than heroic prediction. For most people, participation will not come from perfectly selecting the next AI winner. It may come from disciplined, diversified exposure to productive assets over time rather than standing completely outside the system.
Third, build personal infrastructure. AI is likely to compress the value of routine execution while increasing the value of synthesis, judgment, integration, and the ability to solve complex rather than merely complicated problems.
Most people will not own the core infrastructure of AI.
But being completely excluded from it is a far greater risk than participating imperfectly.
If financial ownership is one side of resilience, productive capability is the other.
You may not own a semiconductor fab. But you can still ask three hard questions:
-
Is all of my income still tied only to my time?
-
Do I own anything that can scale without my constant presence?
-
Am I participating in the systems reshaping the economy, or only paying into them?
Those questions are not a complete strategy.
But they are a beginning.
The deeper shift
The biggest change of the AI era is not just technological.
It is architectural.
For more than a century, industrial and corporate society taught people to treat employment as the central container of security. The AI era does not eliminate work. It destabilises the role that work once played in organising life. It weakens the assumption that effort inside institutions will necessarily translate into durable personal security outside them.
That is why the argument is not simply that everyone should invest.
It is more profound than that.
People need to redesign how they think about security itself.
Not only salary, but ownership.
Not only work, but also leverage.
Not only employment, but systems.
Not only survival, but the ability to buy back time, autonomy, and room for meaning.
In that sense, investing in the AI era is not merely about wealth accumulation. It is about refusing to remain entirely exposed to a world in which value is increasingly captured by ownership, infrastructure, and scale.
Conclusion
The AI era is often described as a race for intelligence.
I think that is only partly true.
It is also a race for position.
A race for ownership, infrastructure, resilience, and leverage.
Jeremy Rifkin warned that technology could erode the centrality of work in modern life. Taylor Pearson, writing from a later and more entrepreneurial generation, pushed the argument into personal strategy: in a world shaped by globalisation, software, and falling barriers to entry, depending on a job may itself become the risk. AI sharpens that reality. It does not merely threaten tasks. It changes the architecture of economic security.
This does not mean despair is the right response.
It means clarity is.
The first step is to stop treating investing as a niche hobby for market enthusiasts. In an asset-driven AI economy, investing is becoming something broader: a way of understanding where power is moving, where value is accumulating, and how not to remain permanently on the labour side of an increasingly asymmetrical system.
In previous eras, not investing may have meant slower progress.
In the AI era, it may mean a structural disadvantage.
In such a world, refusing to think about ownership is no longer neutral.
It is a vulnerability.
About the author
Sinclair Huang is an independent researcher and executive advisor working at the intersection of AI, industrial strategy, semiconductors, and capital allocation. His writing focuses on how technological change reshapes value creation, economic power, and strategic positioning across industries.
Notes and sources
-
World inequality and wealth concentration — World Inequality Report 2026, Executive Summary: <https://wir2026.wid.world/insight/executive-summary/>
-
AI, productivity, and wealth inequality — IMF Working Paper, AI Adoption and Inequality (2025): <https://www.imf.org/en/publications/wp/issues/2025/04/04/ai-adoption-and-inequality-565729>
-
AI adoption, work, and uneven diffusion — OECD, AI and work: <https://www.oecd.org/en/topics/sub-issues/ai-and-work.html>
-
Emerging divides in AI transition — OECD Regional Development Papers, Emerging divides in the transition to artificial intelligence (2025): <https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/06/emerging-divides-in-the-transition-to-artificial-intelligence_eeb5e120/7376c776-en.pdf>
-
Jeremy Rifkin and the “post-market era” / “third sector” framing — Foundation on Economic Trends, The End of Work: <https://foet.org/project/the-end-of-work/>
-
****Taylor Pearson’s summary of The End of Jobs — <https://taylorpearson.me/summary/>
-
J-PAL overview of randomized evaluations — <https://www.povertyactionlab.org/resource/introduction-randomized-evaluations>
-
2019 Nobel Prize in Economic Sciences — official summary: <https://www.nobelprize.org/prizes/economic-sciences/2019/summary/>
Disclaimer
This essay is intended as a strategic and analytical reflection on AI, work, ownership, and economic structure. It is not investment advice, legal advice, or a recommendation to buy or sell any security or financial product. Readers should make decisions based on their own objectives, constraints, and professional advice where appropriate.
Hashtags
#AI #FutureOfWork #EconomicSecurity #Inequality #Ownership #Strategy