Who Will Own the AI Economy?

As artificial intelligence creates enormous new fortunes, America is facing a question that goes beyond jobs: Who should own the wealth generated by the machines transforming the economy?

Artificial intelligence is usually discussed as a threat to jobs. That may be the wrong place to start.

The more consequential question could be simpler: who will own the wealth AI creates?

If advanced AI systems meaningfully increase productivity, the companies building and deploying them could generate extraordinary value. Investors, founders, and technology workers stand to benefit enormously. Millions of Americans who don’t own significant stock could watch the boom from the sidelines.

That possibility has pushed an unusual idea into the political mainstream: giving ordinary Americans a direct ownership stake in the economy, an idea sometimes called universal basic capital, or UBC. Unlike traditional welfare, it isn’t primarily about giving people money to spend today — it’s about giving them assets that can compound over time. That distinction could matter a great deal as AI reshapes the relationship between work, income, and wealth.

From Income to Ownership

The idea is related to universal basic income, but the philosophy is different. UBI provides regular payments; UBC seeks to hand people ownership. In practice, a basic-capital system could give individuals diversified investment accounts that grow in value alongside the broader economy — not to compensate people for disruption, but to give them a stake in the growth itself.

That distinction matters because wages and wealth aren’t the same thing. A person can hold a steady job and still accumulate very little wealth. Someone else can own shares in successful companies and grow considerably richer without a bigger paycheck at all. If AI increases company valuations faster than it increases wages — which is already the pattern in the current AI investment boom — ownership becomes the more important variable.

AI Could Make the Ownership Gap More Visible

America already has a substantial divide in asset ownership. Stocks, retirement accounts, and real estate have let many households ride economic growth, but access to serious financial assets remains highly unequal.

That’s a real problem for the AI era: imagine AI dramatically raising business productivity, company values rising, and entire new industries emerging — an extraordinary economic achievement that, if the gains flow mostly to people who already hold financial assets, could simultaneously widen the gap between wealthy households and everyone else. The risk isn’t necessarily that AI destroys millions of jobs. It’s that AI could generate enormous wealth without distributing ownership of it broadly enough to matter.

The Case for Giving Everyone a Stake

This is the strongest argument for UBC: rather than waiting for inequality to become politically explosive, give people assets now and let compounding do the rest over decades. One well-known version of the idea, proposed by legal scholars Bruce Ackerman and Anne Alstott, would give every American $80,000 at age 21, funded by a wealth tax, to invest, spend on education, start a business, or save — the theory being that a meaningful capital stake at the start of adulthood changes what choices are actually available to someone.

Supporters of broad ownership come from strikingly different political traditions — some see it as a progressive answer to inequality, others as a market-oriented alternative to expanding government welfare. That range of support is itself part of what makes the idea interesting: it doesn’t require agreement about capitalism generally, only about one narrower proposition — that ownership should be spread more widely.

There’s Already a Small Experiment Underway

This isn’t purely theoretical. As part of the One Big Beautiful Bill Act, signed into law in July 2025, the federal government began seeding new “Trump Accounts” with $1,000 for every child born between January 1, 2025, and December 31, 2028, invested automatically in a low-cost U.S. stock index fund and left to grow, tax-deferred, until adulthood.

Families can add up to $5,000 a year, with up to $2,500 of that eligible to come pretax through an employer. A philanthropic gift from Michael and Susan Dell’s foundation is adding another $250 for lower-income children born before 2025. It’s a modest program by UBC standards, but it’s a live example of the broader principle: give people an asset early and let markets do the rest, rather than trying to predict which specific companies will dominate. That distinction — diversified exposure to growth, versus betting on individual winners — could determine whether the larger idea succeeds or collapses under its own complexity.

The Government Ownership Question Is Much Harder

There’s a far more ambitious — and controversial — version of this idea already moving through Congress. Senator Bernie Sanders has introduced the American A.I. Sovereign Wealth Fund Act, which would require major AI companies to transfer a 50 percent equity stake, in a one-time transaction, to a government-managed sovereign wealth fund overseen by a seven-member commission — with the fund gaining voting rights and board seats as it grows, and eventually paying dividends directly to the public.

It’s a striking proposal, modeled loosely on public wealth funds like Alaska’s Permanent Fund or Norway’s Government Pension Fund Global, and legal experts note that affected companies would likely challenge its constitutionality if it advanced.

At first glance, direct government equity looks appealing: if AI companies become extraordinarily valuable, the public participates directly, and the returns could fund public programs or citizen payments. But it creates a real structural problem.

The government would no longer be only the AI industry’s regulator — it would also be one of its owners, with a financial incentive to protect the companies it holds stakes in even when tougher regulation might be warranted. Regulator, shareholder, and policymaker all at once is a genuinely dangerous combination of roles for one institution to hold.

The Risk of Political Control

There’s a political problem layered on top of the structural one. A public fund worth hundreds of billions of dollars — or more — would become enormously powerful in its own right. Who controls it? Who appoints its managers, and how insulated are they from elected officials? What stops politicians from steering investment toward favored companies or industries? What happens when the government owns a meaningful stake in firms whose products raise hard questions about privacy, national security, or employment? These aren’t peripheral details — they’re the whole design problem.

A system meant to democratize wealth could just as easily concentrate power in Washington instead, and advocates of sovereign-fund models generally concede that a poorly governed fund could be worse than no fund at all: it creates the appearance of shared ownership while delivering the reality of politically managed handouts.

There’s also a more mundane limitation worth being honest about: much of the value AI is generating right now sits inside private companies — OpenAI, Anthropic, xAI, and similar firms — that aren’t publicly traded. A broadly diversified sovereign fund built from public-market holdings would likely capture only a fraction, perhaps 30 to 40 percent, of AI-driven wealth rather than the concentrated upside of the specific firms actually driving the boom. That doesn’t kill the underlying argument for UBC, but it does mean expectations for what any near-term fund can realistically deliver should stay modest.

America Doesn’t Have to Choose Between Markets and Public Ownership

There may be a middle path. Policymakers could expand tax-advantaged investment accounts, build on the Trump Accounts model for children, encourage employer contributions, or establish diversified funds that give citizens long-term exposure to growth — all without the government taking direct control of private companies.

The UK has already tried a smaller-scale version of this logic on the supply side: in April 2026 it launched a £500 million Sovereign AI Fund that invests directly in early-stage British AI companies, aiming to build homegrown capability rather than distribute ownership to citizens directly, but demonstrating that state capital and private AI markets can coexist without full nationalization.

A U.S. version focused on broad ownership, rather than industrial policy, would keep the government’s role centered on expanding access rather than managing the companies generating the wealth — preserving the advantages of private markets while addressing how unevenly their gains get distributed.

What Happens If AI Creates Fewer Jobs?

This debate gets more urgent if AI eventually displaces a meaningful share of human labor. Today’s economy still assumes people mainly earn income by working. If machines increasingly perform more of the economically valuable tasks, that relationship weakens — people might still work, but the share of national income flowing to labor, versus capital, could shrink.

In that world, creating more jobs may not be enough on its own; people may need another way to participate in growth, and ownership is one clear answer. Workers who hold productive assets can benefit from technological progress even when that progress reduces the labor required to produce goods and services.

But AI May Not Produce a Dystopia

It’s worth resisting the temptation to build policy entirely around the most pessimistic scenario. Technology doesn’t always destroy jobs permanently — it tends to eliminate some occupations while creating new industries, products, and forms of work that are hard to picture in advance. AI could raise productivity substantially without triggering mass unemployment, which means UBC’s strongest argument shouldn’t depend on predicting an AI-driven job apocalypse.

The case for broad ownership can stand on its own even in the rosiest scenario: if AI makes the economy substantially richer, more Americans owning a piece of that growth strengthens both the economic and political foundations of the country.

The Real Debate Is About Capitalism’s Next Phase

For most of modern American history, the central economic question has been how people earn income. The AI era could push a different question to the center of politics: how do people own the productive capacity of the economy? That’s a considerably bigger question than whether to send Americans another monthly check — it goes to the structure of capitalism itself.

If AI becomes one of the most productive technologies in history, ownership of AI-related assets could become one of the most important sources of wealth for decades, and how that ownership gets distributed will shape inequality accordingly.

A Chance to Build Wealth Before the Crisis Arrives

The advantage of universal basic capital is that it can be pursued before a crisis hits, rather than after. Policymakers don’t have to wait for millions of job losses or politically destabilizing inequality — they can start expanding asset ownership while the economy is still functioning normally, the way Trump Accounts have already begun to do on a small scale. But the design choices matter enormously.

A model built around liquidity, meaningful scale, and some form of ownership representation could genuinely spread the benefits of growth. A model that hands politicians direct control over major private companies risks a very different outcome. That’s the real difference on the table: democratizing wealth, or centralizing economic power under a different name.

The AI Revolution Will Test More Than Technology

AI may eventually transform productivity, employment, and entire industries. But its most important political test may be simpler than any of that: will the gains stay concentrated among people who already own capital, or will a much larger share of Americans become owners of the AI-driven economy themselves?

There’s no guarantee UBC is the right answer — there are serious open questions about funding, investment risk, government involvement, and political independence that current proposals, from Sanders’ sovereign wealth fund to the UK’s more modest venture fund, are still working through in very different ways. But the underlying question is getting harder to ignore: if AI creates extraordinary wealth, who gets to own it? America has a real opportunity to answer that question deliberately, before the technology forces an answer on it by default.