OpenAI has raised a striking idea in its talks around AI policy: give the US government a 5 percent ownership stake. According to the Financial Times, the proposal has been discussed as a way to reduce friction with the Trump administration and respond to public concern over the growing power and wealth of AI companies.
The concept is simple, but the implications are broad. If the government owned part of OpenAI, the public would have a direct financial interest in the company’s rise. At the same time, the arrangement could reshape how Washington approaches one of the most valuable and politically sensitive sectors in technology.
A proposal built around public upside
The Financial Times reported that CEO Sam Altman argued a public financial stake would be the best way to share the upside of AI. The report cited two unnamed people familiar with the talks and said Altman first pitched the idea to Trump early last year.
Altman reportedly suggested the 5 percent figure. Based on OpenAI’s latest funding round, which valued the company at $852 billion, that stake would be worth roughly $42.6 billion.
That number explains why the idea is politically significant. A small percentage of a very large AI company can translate into a major public asset. The proposal also frames AI not only as a private technology race, but as a source of economic value that public officials may want to capture, manage or redistribute.
Still, the talks are described as early-stage. The Financial Times reported that the plan would also involve other US AI companies giving similar stakes to the government. It is not clear whether those companies would accept such a deal.
Why the Trump administration matters
The proposal comes during what the source describes as an unusually hands-on approach to AI from the Trump administration. That approach has already affected OpenAI’s competitor Anthropic and has raised concern over how future interventions could shape the market.
Earlier this year, the Pentagon designated Anthropic a supply chain risk. Last month, the administration unexpectedly placed export controls on Anthropic’s latest models, forcing the company to pull them from the market. The move also created uncertainty about the prospects for US AI on the world stage.
Against that backdrop, a government ownership stake could be seen as more than a financial arrangement. It could also be a political signal. By offering the public a share of AI’s potential value, OpenAI may be trying to reduce the pressure for tougher intervention while positioning itself as aligned with national interests.
The source does not say that any agreement has been reached. It also does not say that regulation would disappear if the government received a stake. What it does show is that ownership, oversight and public benefit are becoming linked in the debate over artificial intelligence.
AI wealth is becoming a policy target
OpenAI’s proposal fits into a wider pattern described in the source: public officials are increasingly interested in using policy to capture and redistribute some of the wealth generated by AI.
Under Trump, the US government has already taken a 10 percent stake in chipmaker Intel. The administration has also reportedly demanded that Nvidia and AMD give the federal government a 15 percent cut of revenue from AI chip sales to China.
Other political figures have gone further in their arguments. Senator Bernie Sanders has argued that AI is a public resource and suggested a one-time 50 percent tax on stock value to create a sovereign wealth fund.
These examples are different in form, but they point toward the same political question: who should benefit when AI companies and chipmakers create enormous new value?
- Ownership stakes give the government a direct financial interest in a company.
- Revenue cuts attach public benefit to specific commercial activity.
- Taxes on stock value aim to convert private market gains into a public fund.
The OpenAI idea sits closest to the ownership model. It would make the government a shareholder rather than simply a regulator or tax collector.
What remains unresolved
The biggest unresolved issue is consent across the industry. The proposal described by the Financial Times would involve other US AI companies giving the government similar stakes. The source makes clear that it is unknown whether they would agree.
That uncertainty matters because a deal involving only one company could create an uneven policy landscape. A broader arrangement would raise even more questions about valuation, governance and the relationship between public officials and private AI firms.
Another open question is how the public would experience the benefit. Altman’s argument, as reported, is that public ownership would share the upside of AI. The source does not detail how the value of a stake would be managed, distributed or used.
For now, the clearest takeaway is that AI policy is moving beyond abstract debates about innovation and risk. The discussion now includes direct claims on the financial value created by leading AI companies.
OpenAI’s 5 percent proposal may never become a final deal. But its existence shows how quickly the AI boom has become a question of public ownership, government leverage and the political price of building one of the world’s most valuable technology businesses.