Trump Weighs Government Equity in AI Firms for Public Fund

President Donald Trump’s recent comments about the federal government taking ownership stakes in leading artificial intelligence firms have drawn attention to an unusual policy idea gaining traction in Washington. The proposal, first pitched to Trump by OpenAI chief executive Sam Altman, would involve major AI labs voluntarily transferring shares to the government, with returns potentially distributed to the public through dividends or other mechanisms. Discussions between the administration and OpenAI have intensified in recent months, though no agreement has been reached.

Trump framed the concept as a way to make the public a partner in the success of these companies. “There are concepts where pieces [of these companies] could be given to the American public,” he said last week. Supporters argue that such an arrangement could address the risk of extreme wealth concentration as AI drives rapid productivity gains. A well-structured public investment vehicle might capture some of those returns and recycle them into broader economic security, echoing models used in certain state-level sovereign wealth funds.

Yet the mechanics remain vague. An informal arrangement limited to a handful of favored firms could easily tilt toward cronyism rather than genuine redistribution. Without clear governance rules, independent oversight, and transparent allocation of proceeds, the approach risks becoming another channel for political favoritism. Past experiments with government equity stakes, from bank bailouts to pandemic relief programs, have shown how difficult it is to balance public returns against private-sector incentives.

At the same time, several of the same AI executives have joined scientists and national security experts in calling for tighter controls on gene synthesis technologies. Altman, along with Demis Hassabis of Google DeepMind and Dario Amodei of Anthropic, signed an open letter urging stronger screening and regulatory standards for commercial DNA providers. The concern is that advanced AI models could lower the barrier for designing novel pathogens, enabling the synthesis of harmful biological agents that would otherwise require sophisticated laboratory infrastructure.

Gene synthesis already supports legitimate medical and agricultural advances, from producing insulin to enabling new cancer therapies. The same tools, however, create dual-use risks that current voluntary guidelines have struggled to contain. Coordinated action among competing AI companies on this issue stands out because those firms rarely align on regulatory questions. Their shared position suggests the threat of AI-assisted bioweapons is viewed as sufficiently serious to warrant industry-wide standards.

The convergence of these two threads—economic partnership and biosecurity regulation—illustrates how AI policy is forcing the Trump administration to confront both the distributional consequences of the technology and its potential misuse. A narrow deal focused only on equity stakes would leave safety questions unaddressed, while new synthesis rules without mechanisms to share AI-driven gains could deepen public skepticism about who benefits from rapid technological change. Policymakers will need to decide whether these issues can be handled through ad hoc negotiations with individual firms or whether they require more durable institutional frameworks.