The Trillion-Dollar Question

Here is a fact that should keep you up at night: we are probably five to ten years away from the first trillionaire. Not in inflation-adjusted dollars, not in theoretical asset valuations, but in real, liquid, spendable wealth. The trajectory is not subtle. Elon Musk crossed $400 billion. Jensen Huang is climbing. The founders and early investors in whatever company solves artificial general intelligence will make the Gilded Age look like a neighborhood yard sale.

The conventional response to this from the left is: tax them. The conventional response from the right is: let them create jobs. Both responses are, I think, catastrophically unimaginative. They are fighting the last war — debating how to divide an industrial-era pie while a fundamentally new kind of wealth is being baked in a fundamentally new kind of oven.

What if the answer is neither redistribution through taxation nor trickle-down through employment? What if the answer is ownership?

The great political failure of the twenty-first century is not that we created billionaires. It is that we created billionaires in companies whose value was built on public infrastructure, public data, and public attention — and gave the public no equity stake in the result.

Consider what went into making OpenAI worth $300 billion. The training data came from the open internet — from blogs and books and forum posts written by millions of people who were never compensated. The foundational research came from universities funded by taxpayers. The chips were designed using decades of publicly funded semiconductor research. The electrical grid, the fiber optic cables, the GPS satellites, the internet protocols — public goods, all of them. And yet the equity sits in the hands of a few hundred investors and employees.

This is not an argument against profit. It is an argument against the exclusivity of profit when the inputs were collective.

The Sovereign Wealth Model

Norway figured something out in 1990 that the rest of the world is still struggling to understand. When they discovered oil in the North Sea, they did not simply tax the oil companies and spend the revenue. They created the Government Pension Fund Global — now worth over $1.7 trillion — and invested the proceeds in global equities. Every Norwegian citizen is, in effect, a shareholder in a diversified portfolio of the world's best companies. The fund owns roughly 1.5% of every publicly listed company on Earth.

The result is instructive. Norway has one of the lowest Gini coefficients in the world (around 0.25), universal healthcare, free education through university, and a sovereign wealth fund that earns more per year than the country's entire oil revenue. They did not redistribute wealth. They created a mechanism for collective wealth accumulation.

The Gini coefficient measures income inequality on a scale from 0 (perfect equality) to 1 (perfect inequality):

The United States currently sits around $G \approx 0.49$ — higher than every other developed nation, and climbing. But here is what is remarkable: the mechanism for reducing $G$ does not have to be taxation. If American citizens held equity stakes in the companies generating the most value, the Gini coefficient would compress not through transfers but through returns.

The last row deserves elaboration. Elon Musk is currently engaged in a legal and political battle over OpenAI's conversion from a nonprofit to a for-profit entity. Whatever you think of Musk's motives — and there is reason for skepticism — the underlying claim has merit: OpenAI was founded with public benefit as its charter, received tax-deductible donations on that basis, and is now attempting to enrich its investors by abandoning that charter. If any portion of that equity were redirected to a public trust, it would create a precedent more significant than the lawsuit itself.

The Mathematics of Collective Ownership

Imagine a People's AI Fund — a sovereign wealth vehicle that holds equity stakes in the major AI companies. Not a majority stake. Not a controlling interest. Something modest: 5-10% of each company, acquired through a combination of regulatory negotiation, public investment, and the leverage that comes from being the government that controls the regulatory environment in which these companies operate.

The compound returns on such a fund are staggering. If the aggregate value of the AI industry grows at even 15% annually (conservative, given current trajectories), a fund with $500 billion in initial AI equity would grow as follows:

At $V_0 = 500B$ and $r = 0.15$, after 20 years:

Divided among 330 million Americans, that is approximately $24,800 per person — not as a one-time payment, but as a growing asset. A family of four would hold roughly $100,000 in AI equity, appreciating at 15% per year. That is not a universal basic income. It is a universal basic wealth — an ownership stake in the most transformative technology in human history.

Capitalism's great insight was that ownership creates alignment. Its great failure was restricting ownership to those who already had capital. The People's Fund corrects the failure without abandoning the insight.

The political beauty of this approach is that it is not socialism. It does not require seizing the means of production. It does not require dismantling markets. It requires the recognition — already implicit in how we talk about AI safety, AI regulation, and the "social license" to operate — that these companies owe their existence to collective inputs and therefore owe a collective return.

The Alignment Problem We Can Solve

There is a grim irony in the AI discourse. We spend billions worrying about aligning artificial intelligence with human values while completely ignoring the alignment problem we could solve today: aligning the economic incentives of AI companies with the welfare of the public that made them possible.

Every dollar of compute that trains a frontier model runs on a power grid maintained by public utilities. Every dataset scraped from the internet traverses infrastructure built with public funds. Every PhD researcher who made a breakthrough in attention mechanisms or reinforcement learning was educated, in part, by taxpayer-funded universities. The public is already an investor in AI. It is simply an investor with no equity, no vote, and no dividend.

The trillionaires are coming. That is not a prediction — it is arithmetic. The question is not whether we can prevent the concentration of wealth. We probably cannot, and attempting to do so through punitive taxation would likely drive the most valuable companies to jurisdictions with friendlier regimes. The question is whether the rest of us are shareholders or spectators.

Norway chose shareholders. America, with its peculiar allergy to collective action and its deep worship of individual wealth creation, has so far chosen spectators. But the window is still open. The AI companies have not yet gone fully public. The equity is still being allocated. The cap tables are still being written.

The people's share is there for the taking. It just requires the imagination to ask for it — and the political will to structure the ask not as a demand for redistribution, but as a claim of co-ownership. We built this, together, whether we meant to or not. The returns should be ours too.