Sam Altman has reportedly floated an unusual idea in his conversations with the Trump administration: give the US government a 5 percent stake in OpenAI, and encourage the other leading American AI companies to do the same. According to the Financial Times, as relayed by the Economic Times and the Times of India, the proposal is still at a conceptual stage, discussed in early talks with President Trump, Commerce Secretary Howard Lutnick, and Treasury Secretary Scott Bessent. It would likely need an act of Congress to become real. But even as a trial balloon, it is a striking thing for the chief executive of a company valued in the hundreds of billions to suggest handing a slice to the state.
The model Altman has reportedly reached for is the Alaska Permanent Fund, which invests the state's oil revenues and pays residents an annual dividend. The analogy is deliberate: it casts AI as a natural resource whose windfall ought to be shared with the public rather than captured entirely by early investors. OpenAI has been circling this idea for a while. In April it proposed a "public wealth fund" to give every citizen a stake in AI-driven growth, and its nonprofit foundation repeated the theme in May, arguing that an AI-shaped economy will need new mechanisms to give people a lasting stake in the systems creating the value.
What makes the timing interesting is who else is talking this way. Altman has reportedly discussed public ownership with Senator Bernie Sanders, who wants the government to hold as much as half of each major US AI company through a sovereign wealth fund. When a democratic socialist senator and a Silicon Valley chief executive are gesturing at the same policy shelf, it is worth asking what each wants from it. For Sanders, public ownership is about capturing profit for citizens on principle. For Altman, the calculus looks more practical.
That calculus is written in the current climate in Washington. AI developers are under mounting scrutiny over data-center expansion, cybersecurity, and job losses. Both OpenAI and Anthropic have seen their most advanced models delayed by government review, and some of Trump's own advisers are pushing for tighter oversight. Against that backdrop, a voluntary equity gift starts to look less like generosity and more like insurance: a way to align the company's interests with the administration's and to defuse the political charge that a handful of firms are about to get very rich at the public's expense. Both companies are also preparing for eventual public listings that would enrich existing shareholders considerably.
The tensions are obvious once you sit with it. A government that owns 5 percent of a frontier AI lab is also the government meant to regulate it, a conflict of interest baked directly into the cap table. A stake distributed through a fund could genuinely spread the gains, or it could simply bind the state more tightly to the fortunes of a few incumbents while smaller rivals get nothing. And the whole scheme presumes the enormous profits it would share actually arrive, which is not yet settled. For now this is a conversation, not a term sheet. But that the conversation is happening at all signals how seriously the industry now takes the question of who should own what AI creates, and how much it would prefer to answer that question on its own terms.