Senior U.S. officials have launched preliminary discussions for the federal government to acquire equity stakes in major artificial intelligence (AI) companies.
The initiative, first reported by digital news outlet NOTUS, stems from a concept pitched directly to President Donald Trump by OpenAI CEO Sam Altman in 2025, with discussions continuing into recent weeks. The proposed framework centers on AI firms voluntarily ceding shares to the government. Returns from these investments would be directed toward public initiatives, potentially including a direct dividend payment to all American households.
The deliberations arrive at a critical juncture for the tech sector. OpenAI is reportedly preparing a confidential initial public offering (IPO), while competitor Anthropic confidentially filed for a U.S. IPO on Monday. While OpenAI has engaged with the administration, a source familiar with the matter noted that Anthropic is not currently participating in talks to provide equity to the government.
Proponents argue that a public dividend could alleviate widespread anxiety over the economic fallout of the AI revolution, ensuring the financial benefits of a technology built on collective human knowledge are distributed broadly. Public skepticism remains high; a recent Quinnipiac poll revealed that 55% of Americans believe AI will do more harm than good, a sentiment that has fueled local resistance to vital data center construction.
OpenAI previously signaled support for this economic model in an April policy paper, which advocated for a Public Wealth Fund to give citizens a stake in AI-driven growth.
However, the proposal faces significant structural and political hurdles. The legal mechanism for a private AI firm to transfer equity to the federal government remains unclear, and critics warn of severe conflicts of interest.
“The problem is that the government would be a shareholder and a regulator at the same time,” said Nat Purser, a senior policy advocate at Public Knowledge, warning that federal oversight could be compromised if safety regulations threaten the government’s own financial portfolio. From the right, Jennifer Huddleston of the Cato Institute cautioned that the move infringes upon traditional free-market principles, arguing it continues a pattern of the government “picking preferred companies.”
Despite concerns, the Trump administration has increasingly embraced partial corporate ownership. The White House recently secured direct investments in at least 10 companies, including a highly touted deal with Intel Corp., and announced a $2 billion equity stake across nine quantum-computing firms in May. Furthermore, President Trump recently signed an executive order requiring leading AI developers to voluntarily submit models for government cybersecurity testing.
The appetite for public ownership spans the political spectrum, though with differing scales of intervention.
Sen. Bernie Sanders (I-VT) called this week for a forthcoming bill that would impose a 50% tax on the stock of elite AI firms such as OpenAI, Anthropic, and xAI and mandate a 50% federal equity stake to place under the control of a sovereign wealth fund.
Former White House chief strategist Steve Bannon similarly urged the administration to reject “tip money” and demand a 50% equity split.
The White House, OpenAI, and Anthropic did not immediately respond to requests for comment. Sources cautioned that planning remains fluid and the unprecedented arrangement may ultimately fail to materialize.
