Sam Altman has reportedly proposed giving Washington roughly $42.6 billion of OpenAI equity at the company's latest $852 billion valuation, framed as part of his long running argument that Americans should share in AI's economic gains.
OpenAI has reportedly offered to give the U.S. government a roughly 5% equity stake in the company, a position worth approximately $42.6 billion at its latest reported $852 billion valuation, as part of a broader pitch by chief executive Sam Altman that Americans should share directly in the economic gains of frontier artificial intelligence. The Financial Times first reported the proposal, with CNBC, the Guardian, CNN, and CNBC Africa confirming the offer within hours. The proposal is not a signed deal, and the White House has not publicly accepted it.
The offer lands inside a specific policy climate. The Trump administration has tightened oversight of frontier AI developers this year, layering export-control directives onto a broader regulatory posture around national security, cybersecurity, and the U.S. race with China. The wire coverage of the proposal frames it explicitly as OpenAI's response to "political blowback" in Washington, including recent restrictions on certain advanced AI models. For OpenAI, that is the kind of regulatory perimeter an equity offer is designed to soften.
Altman's framing has been consistent for months: frontier AI is too consequential to be governed as a purely private commercial project, and the public deserves a financial claim on the returns. He has called the vehicle for that claim a "public wealth fund," an explicit acknowledgement that the proposal is meant to do more than secure favorable regulatory treatment. It is meant to reframe OpenAI's relationship with Washington from a regulated company into a quasi-public enterprise, in which taxpayers hold an equity claim and, by implication, a seat at the table on how AI is deployed. That posture, as relayed by the Guardian, is the stated rationale behind a 5% offer that would, on paper, cost the company roughly $42.6 billion in diluted ownership.
That implication is where the offer gets interesting. A 5% equity stake does not by itself grant the U.S. government any specific governance rights; equity is not the same as a board seat, a veto over model releases, or approval authority over safety thresholds. What it does is create a financial alignment between Washington and OpenAI's commercial success, and, more importantly, it positions the government as a stakeholder rather than a regulator in the public narrative. Whether that positioning survives contact with OpenAI's existing cap table, its nonprofit-controlled governance structure, and its other large investors is the open question the proposal now forces.
A second question is whether other frontier labs would follow. The Financial Times, as relayed by CNN and IBTimes, explicitly reports that it is unclear whether Anthropic, Google DeepMind, or other U.S. AI developers would accept a similar arrangement. OpenAI is the first frontier lab to put this kind of offer on the table publicly, and it has done so in a specific moment of regulatory vulnerability. That makes the offer an experiment as much as a negotiation. If Washington signals interest, the proposal becomes a template. If it is rebuffed, it becomes a record of how far one AI lab was willing to go to buy itself political room.
Follow-up reporting from IBTimes has treated the proposal primarily as a deal story: how much equity, at what valuation, with what conditions. The harder story is governance. If American taxpayers end up holding a meaningful slice of the world's most consequential AI company, the question is not what the stake is worth. The question is what decisions follow from owning it.