OpenAI might be battling Google and Anthropic to retain its market share, but it’s focusing on how the world will prepare for a post-AGI world.
In a sweeping policy paper published this April, the company has proposed creating a Public Wealth Fund that would give every American citizen an automatic stake in AI-driven economic growth — regardless of whether they own stocks or have access to capital markets. The proposal is part of a broader 12-page industrial policy document titled “Industrial Policy for the Intelligence Age: Ideas to Keep People First,” which lays out OpenAI’s vision for governing the transition to superintelligence.
The Fund
The core idea is straightforward: as AI creates enormous economic value, most of it risks pooling at the top. OpenAI’s proposed Public Wealth Fund would invest in “diversified, long-term assets that capture growth in both AI companies and the broader set of firms adopting and deploying AI,” with returns distributed directly to citizens. Think of it as a sovereign wealth fund built specifically for the AI age — one where every American, not just investors, gets a cut.
The proposal doesn’t prescribe exactly how the fund would be seeded, leaving that to governments and AI companies to work out together. But the intent is clear: prevent a scenario where AI’s enormous productivity gains flow only to shareholders while workers and communities are left behind.
Not The First To Float This Idea
OpenAI isn’t alone in proposing this. Anthropic made a similar suggestion in its own economic policy paper published in October 2025. In that document, Anthropic proposed sovereign wealth funds that would allow governments to acquire positions in AI-related assets — explicitly framed as a mechanism to “distribute AI-derived wealth more equitably” in scenarios where the AI sector captures an outsized share of economic output.
Anthropic also cited a related UK proposal: an “AI Bond” that would pool investment in the AI stack and distribute returns across Britain, even as AI research concentrates in a few cities like London.
That two of the most powerful AI companies in the world are independently converging on the idea of public stakes in AI growth is notable — it signals growing anxiety inside the industry itself about wealth concentration as AI begins automating professional work at scale.
The Bigger Picture: A New Industrial Policy
The wealth fund is only one piece of OpenAI’s broader agenda. The paper frames the transition to superintelligence as a civilizational shift comparable to electrification or mass production, and some of its other proposals are equally striking. It calls for rebalancing the tax base away from payroll taxes — which fund Social Security and Medicaid — toward capital gains and corporate income, warning that AI will erode the labor-income base those programs depend on. It proposes a “Right to AI” that treats access to foundational models as a public utility, consciously learning from the failure to universally deploy the internet. And it calls for AI data centers to pay their own way on energy — a pointed position given OpenAI’s own trillion-dollar infrastructure ambitions.
On the labor side, the document proposes converting AI efficiency gains into concrete worker benefits: better retirement contributions, expanded healthcare, and four-day workweek pilots at full pay. The logic is blunt — if AI cuts operating costs, workers should see some of that dividend, not just shareholders. It also calls for formal worker voice in AI deployment decisions, with limits on uses that compress autonomy or undermine fair pay. On safety, OpenAI wants auditing regimes for frontier AI risks, international information-sharing between governments and labs, and governance structures that prevent any individual or internal faction from quietly using AI to concentrate power.
The Credibility Question
It’s worth noting the tension at the heart of this document. OpenAI is simultaneously one of the most powerful forces in AI concentration and the author of a paper warning against concentration of wealth and control. The company recently completed its conversion to a for-profit structure after years of internal debate, and has committed over a trillion dollars in infrastructure spending through partnerships like Stargate.
OpenAI acknowledges it doesn’t have all the answers, and frames the document explicitly as a “starting point for discussion” rather than a final policy blueprint. It’s welcoming feedback at [email protected] and has committed up to $100,000 in grants and $1 million in API credits for policy research building on these ideas.
Whether governments treat this as genuine advocacy or sophisticated lobbying will depend on what follows. But the ideas themselves — a public wealth fund, worker voice in AI deployment, adaptive safety nets, and democratic accountability for AI systems — are worth taking seriously on their merits, regardless of who’s proposing them.
The transition to superintelligence is already underway. The policy conversation is only just beginning.