Bernie Sanders’ $7 Trillion AI Tax: How a Sovereign Wealth Fund Could Reshape Power in Tech

Sen. Bernie Sanders introduced legislation Thursday that would force the largest artificial intelligence companies to hand over half their equity to the American public. The American AI Sovereign Wealth Fund Act, detailed first in an Associated Press report, aims to create a fund valued at nearly $7 trillion at current market prices. It would pay every American more than $1,000 a year. And it would give government-appointed directors seats on corporate boards.

The proposal lands at a moment of intense debate over who should control the future of artificial intelligence. Tech executives warn that heavy-handed intervention could drive innovation overseas. Supporters argue the technology rests on data and creativity generated by millions of ordinary people. Sanders wants those people to get a direct financial stake.

Under the bill, any AI company that generates more than $200 million in annual sales tied to artificial intelligence would face a one-time 50 percent tax paid in stock rather than cash. The government would become a major shareholder. A seven-person independent commission, nominated by the president and confirmed by the Senate, would oversee the fund and exercise voting rights. Those shares could block decisions deemed harmful to citizens and push for policies that deliver broader benefits.

“The benefits cannot simply go to the handful of wealthy corporations,” Sanders told the AP. “They will be shared by the American people.” He added that the public needs “a significant seat at the table to make sure that terrible things do not happen to ordinary people, and that in fact, AI benefits ordinary people, not hurts them.”

The Vermont independent has built the case for weeks. In a June 1 New York Times opinion column he argued that the creative work of millions has been effectively taken by a small group of billionaires. His legislation formalizes that view. A 5 percent annual dividend from the fund would fund the $1,000-plus payments. Additional gains would flow to health care, education and housing programs. Sanders insists taxpayers face no downside if valuations collapse. The government simply holds the equity.

Reactions split sharply along familiar lines. Some conservatives called the plan outright nationalization. Others noted an unexpected alignment. President Trump has floated similar ideas about government equity stakes in AI firms, according to multiple reports. In one AP story on industry talks, Trump said there is “something very interesting” about turning AI into a partnership with the American public. He suggested he and Sanders “aren’t that far apart” on certain economic questions.

OpenAI CEO Sam Altman met with Sanders after the initial proposal surfaced. Altman expressed support for giving the public some equity but stopped short of endorsing a 50 percent threshold. Fortune reported that Altman wants AI to “work for ordinary people” while cautioning against measures that could slow progress. The Hill covered the bill’s unveiling and noted Sanders’ estimate that the fund could generate hundreds of billions of dollars each year.

Industry voices pushed back quickly. On X, Y Combinator CEO Garry Tan described the idea as “a war on building in America.” Others warned of capital flight, reduced incentives for founders, and bureaucratic meddling in fast-moving technical decisions. Bruce Schneier, writing on his security blog, acknowledged the appeal of democratic oversight but questioned whether government board seats would produce better outcomes than market forces. He pointed to Sanders’ core question: should the future of humanity rest in the hands of a few billionaires with little democratic input?

The mechanics matter. The tax applies only to AI-related revenue above the threshold. New entrants would face the same rule once they scale. The commission holds equal representation on boards alongside private shareholders. Voting power could influence everything from model safety standards to deployment choices to labor practices. Proponents see this as necessary guardrails. Critics see a recipe for political interference in technology that already faces regulatory scrutiny on energy use, data centers and national security.

Sanders has paired the ownership push with other AI measures. He introduced the Artificial Intelligence Data Center Moratorium Act earlier this year, seeking to pause new large-scale facilities while impacts are studied. That bill, tracked on Congress.gov, reflects his concern about unchecked expansion. The sovereign wealth fund proposal goes further. It treats advanced AI models as a public resource akin to oil on federal land. The accumulated knowledge, text, images and code scraped to train today’s systems came from society at large, Sanders argues. Therefore society should share in the returns.

Economists and policy analysts have long discussed sovereign wealth funds. Alaska’s Permanent Fund distributes oil revenues to residents. Norway manages its petroleum wealth for future generations. Sanders adapts the model to the digital age. Yet AI differs from finite natural resources. Its value depends on continuous innovation, talent attraction and computational infrastructure. A sudden 50 percent dilution could affect stock prices, investment rounds and executive compensation across the sector.

Recent coverage adds texture. The Washington Post examined how Trump’s interest in public stakes overlaps with Sanders’ plan while noting the details remain fluid. Gizmodo highlighted the $1,000 annual payment and the argument that AI training data represents a collective inheritance. Tangle News outlined the contrasting motivations: Sanders emphasizes redistribution and control, while Trump frames it as strategic national advantage.

But the core tension persists. Can government directors steer AI development without stifling the very breakthroughs that create the fund’s value? Will companies accelerate efforts to keep revenue below the $200 million trigger or relocate key operations? And how might a future administration with different priorities use those board seats?

Sanders’ bill faces long odds in a divided Congress. Yet it crystallizes a growing populist frustration. Polls show many Americans worry about job displacement, concentrated power in tech, and the rapid pace of change. The promise of direct payments taps into that anxiety. So does the language of reclaiming what belongs to the public.

Tech leaders have offered their own versions of sharing gains. Altman once proposed giving citizens equity in AI companies or a national fund funded by industry profits. Those ideas were voluntary and smaller in scale. Sanders makes the transfer mandatory and massive. The difference reveals a deeper disagreement about property rights in the age of machine learning.

Implementation would prove complex. Valuing private AI companies, defining AI revenue, managing conflicts on boards, and insulating the commission from political pressure all present hurdles. Markets would react. Valuations of Anthropic, OpenAI, xAI and other targets could adjust in anticipation. Venture funding patterns might shift toward smaller models or non-AI applications.

Still, the proposal forces a conversation. AI systems learn from human culture, science and expression on a scale never seen before. Their economic impact could dwarf previous technologies. Who captures that value matters. Sanders places his bet on direct public ownership. Others prefer regulation, taxation of profits, or letting competition and markets distribute gains.

The coming months will test these ideas. Hearings, counterproposals, industry lobbying and public opinion will shape what, if anything, emerges from this moment. For now Sanders has put a bold number on the table. Seven trillion dollars. Half the equity. Annual checks to every citizen. Control through ownership. The debate has moved from abstract principle to legislative text.

And that text will face intense examination. Supporters see justice and stability. Opponents see danger to the engine of progress. The outcome could influence not only AI’s trajectory but the broader relationship between government and the technology industry for decades ahead.