WASHINGTON – The most significant thing about OpenAI Inc.’s proposal to donate five percent of its equity to a US sovereign wealth fund is not the percentage. It is what the donation would make the US government.
The Financial Times reported Thursday, citing two people familiar with the matter, that OpenAI chief executive Sam Altman has proposed transferring a portion of the company’s equity to a government-managed investment fund. Other artificial intelligence companies would make similar contributions under the framework. The stated rationale, per the Financial Times’ sources, is to “secure good relations with the administration and address political blowback” – corporate risk management conducted through the mechanism of a gift to the state.
At OpenAI’s January 2026 valuation of $157 billion, five percent would represent approximately $7.9 billion in notional value. The arrangement, if implemented, would make the federal government a direct financial beneficiary of OpenAI’s commercial success: every API call, every enterprise subscription, every licensing deal the company strikes would generate a return for a US government-controlled fund. The incentive structure that creates for the government’s posture toward OpenAI’s competition and regulation has not been publicly discussed.
President Trump offered his own framing for the concept in June, after CNBC first reported early-stage discussions. The administration, he said, was exploring “concepts where pieces could be given to the American public, where the American public essentially becomes a partner with the companies.” The presentation is civic-minded. The mechanics are more specific: a private company proposing to make the government that oversees its regulatory environment into a financial partner.
OpenAI provided the policy scaffolding in April, publishing a paper titled “Industrial Policy for the Intelligence Age” that proposed a public wealth fund for AI investment. “Returns from the Fund could be distributed directly to citizens, allowing more people to participate directly in the upside,” the document stated. The language emphasises citizen distribution. TechCrunch reported the Financial Times’ disclosure Thursday. The two frames – civic benefit and political accommodation – describe the same transaction from different vantage points.
The proposal arrives as OpenAI is mid-transition from a nonprofit research organisation into a for-profit public benefit corporation, a restructuring that converts its accumulated intellectual capital and brand into a conventional equity structure for the first time. A five percent equity donation, in that context, is not abstract: it is a proposal to assign a government fund a direct claim on the future earnings of an entity still determining what its equity actually means and who the other shareholders will be.
The contrast with a more aggressive version of public AI ownership was put on record in June by Senator Bernie Sanders, independent from Vermont. His American AI Sovereign Wealth Fund Act would impose a one-time 50% tax on the stock of companies classified as “systemically important” in the AI sector, with collected shares deposited into a public fund. A 50% stake is ten times what OpenAI is offering voluntarily. The bill has not advanced to committee.
That distance – five percent offered freely, fifty percent proposed legislatively and going nowhere – is the actual range of public AI ownership currently on the table. The voluntary donation is calibrated to generate political goodwill without constraining the company’s operational independence. The mandatory tax would transfer substantially more economic value to public hands. The administration processing OpenAI’s proposal has given no public signal it favours mandatory redistribution over voluntary arrangements that achieve the same political outcome at lower cost to the companies involved.
The calculation extends beyond OpenAI. As this publication reported, Apple is simultaneously lobbying the Trump White House to approve chip deals with Chinese suppliers that the Pentagon has designated military risks – a structurally different transaction reflecting the same underlying logic: managing the administration relationship has become a cost of doing business at scale in the AI economy. The tools differ; the objective does not.
For institutions outside the United States that depend on OpenAI’s infrastructure – hospitals using clinical AI tools, governments deploying the company’s models in public services, businesses whose operations run on its API – a US government equity stake raises questions neither the policy paper nor the reported proposal has addressed. A government holding financial interest in a commercial AI platform operates under different incentives than a regulator or a neutral supplier. What that means for cross-border data commitments, for pricing in non-US markets, or for access in countries whose relations with Washington deteriorate is a question the reporting leaves open.
Any formal arrangement requires congressional approval, and the discussions remain preliminary. This week’s other major intersection of technology and government power moved in the opposite direction: Google definitively lost its final EU appeal over a €4.1 billion Android antitrust fine, a case in which a regulator imposed a financial consequence on a platform company rather than accepting a voluntary financial relationship. The OpenAI proposal runs the logic in reverse. Whether five percent, donated voluntarily on terms the company controls, produces the political stability it is designed to purchase is the question the talks have not yet answered.
