OpenAI Pitches Giving the US Government a 5% Stake
OpenAI has proposed handing the United States government a 5 percent equity stake in the company, a holding worth roughly $42.6 billion at its most recent $852 billion valuation, according to a Financial Times report published Thursday and confirmed by multiple outlets. The pitch, delivered directly by CEO Sam Altman to President Donald Trump and senior administration officials, would route the shares into a new sovereign wealth fund modeled on the Alaska Permanent Fund, with the returns eventually flowing back to American citizens.
The talks remain conceptual, according to the FT, which cited two people familiar with the discussions. Altman has reportedly engaged with Trump, Commerce Secretary Howard Lutnick and Treasury Secretary Scott Bessent on the idea. Any formal arrangement would likely require an act of Congress, a hurdle that could stall or reshape the proposal considerably.
The plan is not limited to OpenAI. Under the version described in the report, other leading American AI companies, a group that would presumably include Anthropic, Google and Meta, would cede similar 5 percent stakes to the same fund. None of those companies has signaled agreement, and it is unclear whether chipmakers such as Nvidia, AMD and Micron, which have captured much of the AI boom's financial upside, would be expected to participate.
Why OpenAI is offering equity to Washington
The timing is not accidental. Per the FT's reporting, the donation is intended to "secure good relations with the administration" and blunt political blowback as anxiety grows over AI-driven job losses. The White House has become an increasingly active gatekeeper for frontier AI. OpenAI's newest GPT-5.6 models launched in late June only to a limited set of users vetted by the US government, and a June executive order established a voluntary framework under which frontier labs give federal officials early access to advanced models before public release.
There is also precedent for the government taking direct positions in strategic technology companies. Last year, Washington acquired a 10 percent stake in Intel as part of an effort to shore up domestic chipmaking. Trump has since floated extending that approach to AI. Speaking to reporters in June, after CNBC first reported that stake discussions were underway, the president described concepts where "the American public essentially becomes a partner with the companies," calling the idea interesting and saying his administration would look into it.
The Alaska model
OpenAI's proposal borrows its structure from the Alaska Permanent Fund, the sovereign wealth vehicle Alaska created in 1976 to invest surplus oil revenue. That fund was valued at roughly $91.2 billion as of May 31 and pays an annual dividend to every Alaska resident. The idea of applying the same logic to artificial intelligence has circulated inside OpenAI for years. In April, the company published a policy paper titled Industrial Policy for the Intelligence Age that called for a public wealth fund investing directly in AI labs and the companies deploying their technology, with returns distributed to citizens regardless of their starting wealth.
OpenAI is not the only party thinking this way, though others want to go much further. Sen. Bernie Sanders introduced legislation in June, the American AI Sovereign Wealth Fund Act, that would impose a one-time 50 percent tax on the stock of systemically important AI companies and deposit the collected shares into a public fund. Companies like Google and SpaceX, where AI is only part of the business, could spin off non-AI units to limit their exposure. The bill has not advanced to committee. Altman has reportedly discussed the broader concept with Sanders directly, an unusual point of contact between the AI industry's most prominent executive and one of its sharpest critics in Congress.
What it means for founders and operators
For startup founders, the headline is not the dollar figure. It is the signal that equity itself is becoming an instrument of AI politics. The largest private company in the world is proposing to dilute its shareholders by 5 percent, not for capital, but for political stability. That reframes government relations as a line item with a measurable cost, and it suggests the biggest players now view regulatory goodwill as something worth tens of billions of dollars.
Three practical implications stand out. First, if a government stake becomes the norm for frontier labs, investors will start pricing political risk into AI valuations well before the growth stage, and term sheets for AI infrastructure companies may begin accounting for potential public ownership. Second, a public fund holding stakes across OpenAI, Anthropic, Google and Meta would give Washington a direct financial interest in the sector's success, which cuts both ways: it could soften antitrust pressure, or it could entrench incumbents whose gains now flow partly to taxpayers. Third, the proposal shifts the AI labor debate. If dividends from an AI wealth fund reach ordinary households, the political argument that AI's gains are hoarded by a handful of companies loses some force, which is precisely why OpenAI is making the offer now.
There is a competitive subtext as well. A company that voluntarily hands the government a stake buys itself a seat at the table where release standards, export rules and procurement decisions get made. Smaller AI startups cannot make that trade, and they should watch closely whether early access programs, federal contracts and export permissions begin to correlate with participation in whatever fund structure emerges.
What happens next
Nothing is imminent. The discussions are early, the mechanics are unresolved, and congressional approval is a significant obstacle in an election-adjacent political environment. Key questions include whether stakes would be donated or purchased, how voting rights would be handled, whether participation would be truly voluntary, and how a fund would treat foreign investors already on these cap tables. The White House is separately finalizing voluntary release standards with OpenAI, Google and Anthropic, and the interplay between those rules and any equity arrangement will be worth watching through the fall.
What is already clear is that the era of AI companies operating at arm's length from the US government is over. The only question is what form the embrace takes, and at what price.
Frequently Asked Questions
What exactly has OpenAI proposed?
According to the Financial Times, OpenAI has held preliminary talks with the Trump administration about contributing 5 percent of its equity to a US sovereign wealth fund, with other major AI companies expected to contribute similar stakes. Returns would eventually be distributed to American citizens, similar to the Alaska Permanent Fund's annual dividend.
How much would a 5 percent stake in OpenAI be worth?
At OpenAI's most recent reported valuation of about $852 billion, a 5 percent stake would be worth roughly $42.6 billion. The actual value would depend on how and when any transfer was structured and on the company's valuation at that time.
Would other AI companies have to participate?
No. The proposal as reported is voluntary, and no other company has agreed to it. OpenAI's pitch envisions peers such as Anthropic, Google and Meta contributing similar stakes, but each would make its own decision, and it is unclear whether chipmakers would be included at all.
Is this the same as the Bernie Sanders proposal?
No. Sanders introduced a bill in June that would impose a one-time 50 percent tax on the stock of systemically important AI companies, a far more aggressive approach than OpenAI's voluntary 5 percent contribution. That bill has not advanced to committee.
Why does this matter for startup founders?
It signals that political risk is becoming a priced cost in AI, that government relationships may increasingly shape access to contracts and release approvals, and that the largest labs are willing to spend equity to secure regulatory goodwill. Founders building in AI should expect investors to weigh policy exposure more heavily in valuations and deal terms.