Senator Bernie Sanders is once again advancing a sweeping government intervention that would fundamentally reorder the relationship between the state and private enterprise, this time by demanding that Washington seize half the equity of Americas largest artificial intelligence firms.
According to the Washington Free Beacon, the Vermont socialistlong aligned with Rep. Alexandria Ocasio-Cortez and New York City councilman Zohran Mamdani as a standard-bearer for the activist lefthas unveiled a plan that would give the federal government a 50 percent ownership stake in major AI companies. The proposal, framed as a populist bid to ensure that the public shares in the gains of a booming sector, would in practice amount to a massive expropriation of private shareholders and a dramatic expansion of federal power over one of the most dynamic areas of the U.S. economy.
"I will soon be introducing the American A.I. Sovereign Wealth Fund Act. This legislation would give the public a direct ownership stake in the largest A.I. companies in our country," Sanders declared in a New York Times opinion piece. "It would create a sovereign wealth fund through a one-time 50 percent tax not on the profits of OpenAI, Anthropic, xAI and other companies, but paid with something far more valuable than that: the stock."
Sanderss plan arrives alongside a May 27, 2026, Time magazine article by Senator Elizabeth Warren titled "Why We Need to Tax AI," underscoring how aggressively leading Democrats are now circling the AI sector. The timing is no coincidence: as AI valuations soar and new billionaires emerge, progressive politicians appear increasingly eager to tap the industry as a revenue source for long-standing redistributionist ambitions.
For the moment, with Democrats in the minority in both chambers of Congress, these proposals function more as ideological markers than imminent legislative threats. An "Artificial Intelligence Data Center Moratorium Act" that Sanders introduced in March has failed to attract even a single Senate cosponsor, a telling sign of how extreme his approach remains even within his own party.
While Ocasio-Cortez was prominently featured in a Sanders press release touting the moratorium bill as a way to "enact a reasonable pause to the development of AI to ensure the safety of humanity," she has yet to introduce companion legislation in the House. Congresss own legislative database shows no House version of the measure, suggesting that even some progressives are wary of being tied to a full-scale regulatory freeze on a transformative technology.
Yet these bills and op-eds still serve as revealing windows into the instincts and priorities that would likely guide Democrats if they regain unified control of Washington after the midterm elections. They point toward a future in which innovation is treated less as a driver of prosperity and more as a convenient target for confiscatory taxation and state control.
Sanderss AI scheme also illustrates how left-leaning academics and foundations are feeding ever more radical ideas into the political bloodstream. The senator openly credits two law professorsJeremy Bearer-Friend of George Washington University Law School and Sarah Polcz of UC Davis School of Lawwith inspiring his proposal, underscoring the pipeline from progressive legal theory to legislative wish list.
Bearer-Friend is not merely an academic; he is a former aide to Warren and a contributor to the Roosevelt Institute, a think tank bankrolled by the Hewlett Foundation, George Soross Open Society Foundations, and the Ford Foundation. That institute previously published a Bearer-Friend paper advocating a $1 trillion tax to fund racial reparations, in which he wrote, "As with any new trillion-dollar tax, we should expect litigation. But the power to tax has been a part of our Constitution since our founding and is not on its face legally dubious."
On December 30, 2025, Bearer-Friend and Polcz published an article in the Columbia Journal of Tax Law outlining "a unique in-kind tax payment structure that would require firms with ownership of AI to remit equity shares to the public." The authors declined to specify a concrete rate, conceding, "The actual rate to be adopted will largely be a political question."
Even so, the article makes clear that the desired rate is not modest. The ideal level, they argue, "will provide the adequate level of public voice in corporate decisions to address public harms of AI" and "will provide a sufficient profit interest to compensate injured workers and creators," while only "minimally" crowding out private investment.
In April, Polcz promoted a popularized version of the law review article in The Hill, a Capitol Hill-focused outlet often used by academics whose submissions are passed over by more prominent publications. "Know any lawmakers looking to tax AI? Weve got a proposal for them," she wrote on social media, candidly advertising the piece as a ready-made blueprint for ambitious politicians.
The current rush to tax AI appears driven less by carefully calibrated policy analysis than by a crude version of the logic attributed to bank robber Willie Sutton: "Because thats where the money is." Bloomberg News recently reported the emergence of 19 new AI billionaires in the past year alone, with a combined net worth of $59.3 billion, making the sector an irresistible target for those who view private wealth as a public resource to be harvested.
Anthropic, the company behind the Claude AI tool, has become a prime example of this explosive growth. On June 1, 2026, the firm announced that it had "confidentially submitted a draft registration statement on Form S-1 to the U.S. Securities and Exchange Commission for a proposed initial public offering of our common stock," following a May 28, 2026, announcement that it had raised $65 billion at a staggering $965 billion post-money valuation.
The Columbia Journal of Tax Law article identifies two primary goals for its proposed equity seizure: "compensating individuals for stolen data and workforce displacement." Both rationales, however, raise serious questions about fairness, effectiveness, and the broader economic consequences of such an unprecedented intervention.
On the data front, many of the entities claiming their information has been "stolen" are hardly struggling victims. New York Times publisher A.G. Sulzberger, a fifth-generation heir to the Ochs-Sulzberger media fortune, recently delivered a speech on what he called AIs "original sin""a brazen theft of intellectual property that has occurred at an unprecedented scale. Tech giants strip-mine news websites without permission or compensation. They repackage these stolen goods as their own, siphoning off the audiences and revenue that otherwise would go to the news organizations that created this work. And this happens not just once during the training process, but countless times every single day."
Sulzberger went on to describe the Timess legal offensive against leading AI players. He noted that "the Times took OpenA.I., its partner, Microsoft, and later Perplexity, to court, for brazen violations of our intellectual property rights But lawsuits are slow and expensive ours has already stretched two-and-a-half years and cost over 20 million dollars."
From the perspective of large media corporations, it is far less "slow and expensive" to enlist sympathetic lawmakers to seize AI company stock and channel the proceeds into compensation schemes. Such an approach would conveniently bypass the burdens of litigation while also preserving a lucrative side channel through which publishers and authors can reward politicians via book "advances" that skirt typical ethics concerns.
The second justification"workforce displacement"is equally fraught. Recently laid-off tech workers may not enjoy Sulzberger-level wealth, but they are generally highly skilled and comparatively well-positioned in the labor market, especially in a growing, innovation-driven economy.
The best environment for such workers to find new opportunities is one characterized by strong property rights, robust private investment, and rapid growth, not a regime in which the government arbitrarily confiscates half the equity of successful firms. When asked what makes AI so uniquely dangerous as to justify a 50 percent equity tax for workforce displacement, Professor Polcz did not respond by deadline, leaving a host of obvious comparisons unanswered.
If AI merits a 50 percent equity seizure to compensate displaced workers, why not impose a similar tax on electric vehicle manufacturers to pay gas-station attendants whose jobs may vanish. Why not levy a 50 percent tax on word-processing software to support former typewriter repairmen, or on solar and wind companies to subsidize coal and oil workers whose livelihoods are threatened by green mandates.
The absence of a clear limiting principle raises the risk of creating a permanent class of self-identified "displaced" workers, incentivized to claim victim status rather than adapt and reskill. History offers a cautionary parallel in the United Nations agency established to support Arab "refugees" after the creation of Israel, where the existence of open-ended compensation has arguably entrenched, rather than resolved, displacement over generations.
Sanders is remarkably candid about his broader ambitions for the revenue he hopes to extract from AI firms. "The proceeds would be used to ensure that every man, woman and child in our country has a decent and dignified standard of living, including health care, education and housing," he has said, effectively presenting the AI sector as a piggy bank for a sweeping expansion of the welfare state.
What this vision ignores is that artificial intelligence itself holds enormous promise to improve precisely those areashealth care, education, and housingthrough innovation rather than redistribution. Robot-assisted construction could dramatically reduce building costs, while self-driving trucks could speed deliveries and lower prices for construction materials, making homes more affordable without new federal entitlements.
Yet Sanders consistently emphasizes the potential downsides of AI while downplaying or dismissing its benefits. "Left unchecked, it is likely that millions of truck drivers, bus drivers, taxi drivers and rideshare drivers will lose their jobs in the next decade," he warned in an April 16, 2026, Fox News opinion piece.
He continued with a stark prediction about employer behavior. "Why would any business want to hire a human worker when it can install AI and robotics and cut its labor costs by 80 to 90%? AI and robots dont take a salary, need a vacation, require health care, or form a union. They just keep working 24/7."
The Columbia paper, in a footnote, goes even further in its catalogue of alleged AI harms. It claims that "environmental devastation, national security risks, and tyrannical state surveillance are all noted harms of the growth of AI," painting a picture of technological progress as an almost unmitigated threat.
At the same time, the authors insist that their proposal is not a step toward socialism, but rather an extension of capitalism. They argue that "by positioning the state as a shareholder in private enterprise the proposal further expands capitalist systems rather than substituting them with Communist ones," a characterization that will strike many conservatives as a semantic dodge for what is, in substance, a massive state grab.
Taken together, these arguments reflect a fundamentally dystopian view of technological advancement, one that sees innovation primarily as a problem to be managed and monetized by the state. Yet there is an irony in how quickly Sanders himself appears to be adjusting his posture as the AI sectors value becomes undeniable.
Less than three months ago, he was championing a bill to "pause" AI development through a moratorium on data centers, effectively trying to slow or halt the industrys growth. Now he is proposing that the federal government own half of the very companies he previously sought to restrain, a shift that suggests he recognizes that "that half is worth a lot more without a data center moratorium than with one."
For conservatives, there is at least one potential silver lining in Sanderss audacious proposal. His call for "a one-time 50 percent tax" to seize a direct ownership stake in AI firms may provide Republicans with a ready-made rejoinder the next time critics portray the Trump administrations "golden share"-style arrangements with companies like U.S. Steel or Intel as unprecedented authoritarian shakedowns.
If the progressive left is now openly advocating that Washington take half the equity of entire industries in the name of social justice and redistribution, it becomes harder to argue with a straight face that more limited, targeted government stakes are beyond the pale. The real debate, then, is not whether the state should ever hold shares, but whether America will embrace a model of limited, rules-based engagementor slide toward the kind of sweeping, punitive confiscation that Sanders and his allies are now eager to normalize.
Login