Elon Musk has argued that there is one and only one path to universal prosperity: AI and robotics. Several other renowned tech thinkers have argued similarly, or at least shared their belief that AGI (Artificial General Intelligence) will bring abundance and drastically reduce the need to work for a living. It's an inspiring claim, and one that matches the seductive vision of a future where machines do most of the work and abundance becomes the norm.
The relevant technology we are talking about here is not the industrial robots we’ve had for decades -specialized machines in controlled environments executing pre-defined routines. Japan is full of those. What changes everything is the emergence of intelligent robots: embodied systems powered by AGI, the great milestone that AI is currently pursuing, that can perceive, reason, learn, and act across a wide range of real-world tasks. Unlike traditional automation, these robots resemble general-purpose labor, which means their economic impact is not marginal; it is structural.
And that is exactly why intelligent robots will not necessarily make everyone wealthy.
The outcome will depend far less on the technology itself than on one fundamental question: who owns the intelligent robotic labor? Productivity will rise, but whether prosperity is shared -or concentrated- will depend on ownership and institutions.
To see why, consider four plausible futures.
1) An Oligopoly of Intelligent Robot Owners
In the first scenario, a small number of corporations or wealthy individuals own most intelligent robots. This is arguably the default trajectory of industrial capitalism: scale advantages and capital intensity naturally push toward consolidation.
If intelligent robotics follows this path, the result could be socially disastrous. A handful of organizations would control not only production, but also logistics and increasingly the real economy’s operating system. Wealth concentration would accelerate, and social instability would become a feature -not a bug- of the system.
It would also create a new and unusually difficult challenge for antitrust regulators. Traditional frameworks were built for industries where market power is visible through pricing and consumer harm. Intelligent robotics breaks those assumptions. A company that owns fleets of intelligent robots may not “raise prices” -it may simply outcompete everyone by operating at near-zero marginal cost, swallowing entire sectors before regulators can even define the market it dominates. Worse, intelligent robotic infrastructure collapses the boundaries between industries: the same platform could manufacture goods, deliver packages, build homes, harvest crops, and operate warehouses. Regulators would have to police not just monopoly power, but monopsony power (control over labor markets), dominance over AI models and data, and ownership of the physical logistics networks that effectively become the economy itself. In this world, antitrust starts to look less like regulating consumer markets and more like regulating utilities.
2) State Ownership of Intelligent Robots
In the second scenario, governments own and operate major intelligent robotic infrastructure. This could better control concentration risks and limit the worst social consequences.
But it comes with major trade-offs: reduced innovation, less dynamism, and a massive shift of power toward the state, raising uncomfortable questions about governance, surveillance, and individual autonomy. It may reduce instability, but it also increases the stakes of political control.
3) Distributed Ownership of Intelligent Robots
The third scenario seems the most desirable: intelligent robots widely owned by individuals and small organizations, democratizing productivity gains, similar to what happened with cars, appliances, and personal computers.
But intelligent robots may not democratize as smoothly. Cars largely do the same job. The model you buy affects comfort and status, but it doesn’t fundamentally determine whether you become rich or poor.
Intelligent robots will be different. The gap between a basic robot in the hands of a low-skilled person and a highly advanced one in the hands of a high-skilled person will translate directly into a huge productivity gap, and therefore a wealth gap. People and firms that can afford and effectively exploit the most capable robots won’t merely gain convenience; they’ll gain leverage. Access to intelligent robotic labor could become the new dividing line, compounding inequality even in a world of “distributed” ownership.
4) Autonomous Communities of Intelligent Robots
The fourth scenario sounds like science fiction but is not impossible: a self-reproducing, self-maintaining community of intelligent robots that interacts with humans for business purposes, but also competes with them across industries and tasks, eventually behaving like a parallel economic civilization.
Even if this scenario is unlikely in the near term, it’s a reminder that intelligent robotics is not merely an economic story; it’s also a governance and security story.
Less Human Labor… and the Demographic Wild Card
Across all scenarios, one conclusion is consistent: the intelligent robotic economy will require far less human labor. That triggers fears of mass unemployment, but there is a countervailing force already in motion: sharply declining birth rates.
Many societies, particularly in East Asia and parts of Europe, are facing shrinking working-age populations. In that context, intelligent robots may not displace workers so much as replace workers that won’t exist. Demographics could soften the transition—at least in countries where population decline is most pronounced.
China as the Robotics Laboratory
If intelligent robotics creates a new economic order, China may become the most consequential testing ground.
China is effectively starting in a version of scenario 2: centralized power enabling rapid industrial policy, regulation, and reallocation. That gives China two advantages: it can capture the experience curve early, and it has strong incentives to prevent economic hardship from igniting social unrest. In short, China may navigate the transition more deliberately—while much of the world, especially liberal democracies, could face a harsher and more chaotic adjustment.
The Real Question: Not Wealth, but Distribution
Intelligent robotics will almost certainly expand productive capacity. But that does not guarantee universal prosperity. Recent research suggests AI could amplify inequality—not necessarily by eliminating jobs overnight, but by shifting bargaining power, concentrating the returns to capital, and disproportionately rewarding those who own the most scalable systems. The IMF (International Monetary Fund) has warned about labor disruption and rising inequality risks, while the OECD has tracked how occupational exposure to AI may reshape wage structures over time. Early empirical assessments, including the Yale Budget Lab’s review of post-ChatGPT labor market trends, suggest the broad effects are still emerging; yet the distributional question is already visible: who captures the productivity dividend first, and who bears the adjustment costs?
So yes: intelligent robots may create abundance. But they won’t automatically create shared abundance.
That is why the central policy challenge of the next decades may be simple to state, and difficult to solve: whether societies can design mechanisms -robot ownership models, access rights, dividends, or antitrust frameworks- that democratize the ownership of intelligent robotic labor before it concentrates beyond repair.
