The 2028 United States presidential elections are shaping up to be the first AI elections. They will be defined by how candidates propose to tackle the transformation triggered by the widespread integration of artificial intelligence (AI) into society.
Fueled by trillions of dollars in investments that are inflating the stock market, the rapid advance of AI is already beginning to reveal the scale of productivity gains, wealth creation, and job destruction that the future will bring. Across the nation, numerous communities are pushing back against data centers that drain their energy supplies. Frontier models as powerful as Mythos, developed by Anthropic, are regarded as so dangerous for their ability to identify software vulnerabilities and potentially facilitate massive cyberattacks that, for now, are restricted to a small circle of trusted companies and governments.
As a result, the concerns raised among the general public have entered what used to be the exclusive domain of Silicon Valley’s tech bros and the venture capitalists who set the pace of innovation.
“The rapid advance of AI is already showing the scale of the productivity gains, the creation of wealth, and the destruction of jobs that the future will bring”
Even recent college commencement speeches that alluded to the benefits of AI have been met with loud boos from graduates who are excluded from entry-level jobs that AI can perform better, without taking on the full student debt needed to reach that point. Pope Leo XIV has joined the debate with his encyclical Magnifica Humanitas, in which he calls for the “disarmament” of AI to ensure it remains human-centered and under human control. State governments across the ideological spectrum, from California to Florida, are beginning to introduce their own regulations amid a vast federal regulatory vacuum.
Regulation and Distribution
The coming years’ debate will focus on two arenas: regulation and distribution.
Out of caution, the Trump White House has called for voluntary assessments of models like Mythos before they are publicly released. That prudent measure continues to meet opposition from tech accelerationists, who argue that it would give China a competitive and geopolitically delicate edge if similar restrictions do not exist.
Therefore, AI safety protocols must be treated as a global public good that the two AI superpowers must agree on and respect. Otherwise, neither China nor the United States can feel secure, and both will seek advantage. This was a point raised by pioneering AI scientists from both China and the West when they convened in 2024 at the Berggruen Institute’s Casa dei Tre Oci in Venice.
“AI innovations are increasingly decoupling productivity and wealth creation from employment and wages”
As for distribution, it is increasingly clear that %AI innovations are decoupling productivity gains and wealth creation from employment and wages. Analysts at the Bank of America Institute have found that the recent productivity gains are accumulating as corporate profits, while worker incomes consistently fall as a share of U.S. GDP.
As we have often written in Noema, the value created by intelligent machines flows primarily to tech magnates who “own the robots” and to the top 10%, who hold 93% of all shares. Meanwhile, the value of work—and its bargaining power to claim a piece of the pie—declines rapidly. In fact, the public, whose data are exploited to train AI models, typically has no rights to the new wealth generated from their information.
Recently, this realization has led to a proliferation of proposals from Democrats, Republicans, and major tech firms to share AI’s enormous gains more broadly. What changes is the means of achieving this.
These proposals fall into two broad camps: income redistribution and wealth predistribution. The former contemplates higher taxes on AI companies and tech billionaires to redistribute their outsized returns toward healthcare and social services, or toward a universal basic income to compensate for widespread job destruction.
The latter, predistribution, envisions policies that ensure that the entire citizenry takes part from the outset in owning the AI-driven economy, instead of attempting to correct inequality merely by redistributing income a posteriori. We have called this “universal basic capital.” The aim is not only to reduce wealth concentration at the top but to build it from the ground up.
Shifts in Political Winds
A sweeping shift in public sentiment, one that has been emerging for months, seems to be solidifying.
As of the summer of 2026, the MAGA coalition that Donald Trump mobilized to win the last election and capture Congress appears on track for a setback in the midterm elections.
“Disaffected youths, Latino citizens, white working-class farmers, and white working-class independents who helped him reach the threshold of victory are deserting in droves”
Trump’s polling support has plunged to primarily his most ardent base, around 34% of the electorate. Disaffected youths, Latino citizens, farmers, and white working-class independents—who felt abandoned and who helped him reach the victory threshold—are deserting en masse. They fault rising living costs, failed foreign ventures like the Iran conflict, and erratic self-promotion gestures such as campaigns to place the president’s image on a $250 bill. With awakened extremism and immigration now tempered, if not quelled, they are seeking answers to their immediate concerns elsewhere.
This situation foreshadows a real possibility that Democrats will seize control of the U.S. Congress and constrain the maneuvering room of an outgoing Trump. That legislative majority is likely to focus on containment or mitigating damages, rather than constructing positively for the future.
Once the ground is cleared, the opportunity will arrive in the 2028 presidential race. As Trump’s decline persists and pulls his successor along, the Democratic Party will be the space where the action concentrates.
Redistribution vs Predistribution
The Democrats’ erratic wandering through a political desert since MAGA’s rise has found a new foothold in AI-driven challenges.
Among Democrats and their allies, a heterogeneous set of redistributive proposals is emerging. A billionaire tax plan floated in a California ballot drive supported by unions would apply a 5% wealth tax primarily on tech billionaires to offset cuts in the state’s Medicaid program. Massachusetts Senator Elizabeth Warren, a consumer advocate and politician, has introduced a plan to reform the tax code so that the tech giants and the ultra-rich “pay their fair share” and fund universal healthcare, universal education, and unemployment insurance for workers displaced by AI.
“Taxing AI is a way to ensure AI gains benefit all Americans, rather than flowing to a small handful of the ultra-rich,” Warren recently wrote in Time magazine. “If millions lose their jobs to AI, we’ll need funds for universal healthcare so that these workers aren’t bankrupted by medical care. If AI reshapes the future of work, we’ll need free education, retraining, and a new job guarantee so every American has well-paid work. And as workers rebuild, we’ll need revenues to strengthen unemployment insurance and keep families afloat. The only way to get there is to reform our tax code.”
“Right now, employers pay payroll taxes on their workers but get tax breaks for investing in technology: effectively, a penalty on hiring humans and a subsidy for buying equipment. In an AI world, that means our tax code incentivizes corporations to lay off people and replace them with AI. That’s wrong. We must level the playing field by raising taxes on corporations and capital gains and closing corporate tax loopholes. One way to address those loopholes? Strengthen the minimum tax for mega-corporations.”
The starting point, he argues, is a special tax on large data centers to cushion the rise in energy costs for consumers.
On the predistribution side, Vermont Senator Bernie Sanders, a leading figure of left-wing populism, is drafting legislation to create an AI wealth sovereign fund “via a one-time 50%” tax, not on the profits of OpenAI, Anthropic, xAI, and other firms, but paid with something far more valuable: their shares, he wrote in a The New York Times op-ed.
“In an AI world, our tax code incentivizes corporations to lay off people and replace them with AI. That’s wrong”
“If enacted,” Sanders writes, “this legislation would do two crucial things. First, give the public a direct say in shaping AI’s future. The future of AI and the transformation of human life it will bring would no longer be dictated by a handful of tech oligarchs. The federal government would have power, through its voting rights and equal representation on each company’s board, to block decisions that harm our citizens and to push policies that help them.”
“Second, this legislation would ensure that the trillions, if not quadrillions, of dollars potentially generated by AI are used to improve everyone’s lives, not merely to enrich the richest people in the world. If AI-driven companies continue growing as quickly as many analysts expect, the value of the sovereign wealth fund will also grow, and the benefits for the American people will grow along with it.”
In Sanders’s view, “the billions, if not trillions, of dollars generated by this fund would provide direct payments to the American people. And as the fund creates more wealth, the returns would be used to ensure that every man, woman, and child in our country enjoys a decent standard of living, including healthcare, education, and housing.”
As the senator notes, even the largest tech firms, like OpenAI, have themselves proposed a ‘public wealth fund that provides every citizen’—including those who do not invest in financial markets—a stake in the AI-driven growth. Anthropic has also called for a ‘national AI equity fund’.
OpenAI CEO Sam Altman has proposed funding the fund with a 2.5% tax on companies that surpass a given market capitalization threshold, payable by transferring shares.
The tech magnate regards that tax as not more inhibiting than the payroll taxes that all companies already pay as a cost of doing business. It would be a way to capture for society as a whole the productivity gains from AI-powered machine intelligence rather than from human labor. Altman proactively pressed for a meeting with Sanders about the senator’s proposal, aligning with his general stance but arguing that a 50% stake would be too extreme.
“Altman proactively pursued a meeting with Sanders about the senator’s proposal, aligning with his broader stance”
Responding to Sanders, President Trump—who had previously floated the idea of a national sovereign wealth fund—gave an open and positive reception to the possibility of the public owning stakes in AI companies. “There’s something very interesting about it,” he told reporters last Friday, “where it almost becomes an alliance with the American public.” He added that the base that supports him and Sanders “is not that far apart”.
Texas Senator Ted Cruz has also weighed in. He agrees with the idea of a public wealth fund but cautions against direct government ownership of AI firms and suggests that citizens’ accounts take the form of index funds or investment funds. Other influential Republicans argue that any transfer of AI company shares to a public fund should be voluntary.
In California, Governor Gavin Newsom, long considered the favorite for the Democratic presidential nomination at this early stage, has championed the idea of a “data dividend” and “universal basic capital” through a wealth fund in which individuals and families—rather than the government—own stakes in the growing AI economy. His CalKids program is an embryonic form of CBU. It provides a public deposit of up to $1,500 in a savings account for every low-income first-grade student in public schools, which is professionally invested to generate compound returns until it can be drawn upon for college—an early-stage universal basic capital for the next generation.
“The distinction between redistribution and predistribution is decisive and will determine whether wealth is shared more broadly among all”
Newsom, who also implemented California’s first state regulations on frontier AI models and opposes the billionaire tax, argued at a recent Center for American Progress forum that Democrats must go beyond tinkering and promote ideas like universal basic capital. Even if Anthropic’s CEO, Dario Amodei, is only partly right in noting that AI will replace around half of white-collar jobs, Newsom says we are heading toward worker precarity.
Like Warren, he contends that the payroll tax of the industrial era incentivizes replacing labor with machines, while capital is subsidized through tax deductions. “The rapid advance of AI,” he warned, “will blow up the current order”.
“Now the blue-collar worker resembles the twenty-something white-collar workers I see in San Francisco, wondering why they aren’t called back after an interview: they start sounding alike,” Newsom said. “That’s a new coalition.”
The distinction between redistribution and predistribution is decisive and will determine whether wealth is shared more broadly among all or whether only income at the top is redistributed, leaving the structural inequality intact.
Trump Has Laid the Groundwork
It is to be expected that the sharpest minds within the Democratic coalition will craft a compromise that preserves elements of redistribution to help correct existing inequalities, while acknowledging that a fair AI-powered economy’s future must involve a predistribution of wealth ownership. Assets ought to remain directly in individual and family accounts, not administered by the government.
To reach that compromise, Democrats should build on what the Trump Administration and the MAGA Congress have already advanced.
Trump’s Money Accounts for Growth and Advancement program—which provides a $1,000 deposit into a personal account for every child under eight, growing over time with market-based compound returns—could serve as a foundational platform for a broader form of universal basic capital as a cornerstone of the social contract in the AI era.
“The time has come for a complete restructuring of the political economy that matches the scale and scope of AI disruption”
As Newsom argues, AI is reshaping the world so profoundly that tinkering is no longer sufficient. The moment calls for a comprehensive restructuring of the political economy that matches the scale and reach of AI disruption and guides its path toward the common good. As the next presidential elections approach, the American citizenry seems clearly ready for this idea; the moment has arrived.
