How can AI wealth be shared with all Americans

Bernie Sanders’ latest proposal that the public should own half of AI is unlikely to become policy anytime soon, but it reflects a broader debate gaining momentum among economists, technology researchers and policymakers: How can Americans benefit if AI creates trillions of dollars in new economic value? There is no shortage of ideas, many untested. However, the risks and increasing public opposition to artificial intelligence make this question important.

AI wealth has rapidly accumulated in the stock market, but many Americans are still limited in how much they can benefit from this growth. Recent survey studies show that a majority of US workers now want to hold companies more accountable through an AI sovereign wealth fund. There are unconfirmed reports that OpenAI is discussing offering a 5% equity stake to the government ahead of its highly anticipated IPO. Meanwhile, Jeff Bezos recently told CNBC that his best policy idea to level the economic playing field is to eliminate federal income taxes for the bottom half of U.S. earners.

Recent survey data shows that this issue is part of a rapid shift in public sentiment towards AI. One Emerson College survey A survey released this week found that only 27% of Americans support building data centers in or near their communities, while 63% oppose it. Public sentiment has deteriorated significantly in less than a year. In a similar survey conducted in December 2025, 33 percent of participants said they would support such developments, while only 42 percent were opposed. Many Americans feel they have nothing to gain and everything to lose from AI.

“When I see the data center proposal, I don’t see progress,” said Northeast Ohio resident Will Hollingsworth, speaking at an April public comment session on a proposed 257-acre data center campus in Portage County. “I see a gamble where big tech companies get the gold while Portage County picks up the tab.”

“We are being asked to sacrifice the lifeblood of our city so that a trillion-dollar company can shave even a penny off its margins,” Hollingsworth said in comments that went viral. he said. “We are being asked to empty our reservoirs [that] a chatbot can write a poem or something [that] Our sheriff can take a picture of you standing next to Bigfoot.”

There are numerous proposals among economists, tech industry researchers, and public policy experts responding to Hollingsworth’s point that the potential outcomes from AI development are heavily skewed in favor of corporations. These include partial public ownership models and other shared equity mechanisms in AI.

Jaron Lanier, a computer scientist who now runs the Office of Chief Technical Officer and Chief Combining Scientist at Microsoft Research, has advocated for a model sometimes referred to as “data dignity,” in which people receive compensation for knowledge and contributions that help build artificial intelligence systems.

“I spent some time with Senator Sanders when he visited the AI ​​community at Stanford,” Lanier said. “One of the two [his proposal] “Whether it’s a good idea or not depends on the nature of the government that is responsible for directing the benefits to people,” he said. If the government were to simply become “another AI company,” he would opt for what he called a more “distributed economic model.”

“Good data and control” can result in enough real money to have a significant impact on people’s lives, Lanier says.

“But if the future is going to be normative Silicon Valley, where people are fictionally rendered useless because their contributions are anonymized and dismissed in favor of pretending that AI does all the work, then some kind of support had better come from a government structure with a participatory/democratic element,” Lanier said.

Paying people directly for AI training data

The challenge is to figure out how to make such a system work. AI models are trained with massive amounts of information from millions or billions of sources. Determining which individual contributions create value and how much they should be paid could face the same criticisms as dogged efforts to compensate people for their search histories; While the sums may be very large in aggregate, the economic value of any individual individual’s data is low.

Raul Castro Fernandez, an assistant professor of computer science at the University of Chicago, recently written About how to fairly compensate the public for AI, refutes The argument is that it is not possible to accurately track (and compensate for) the massive amounts of data points collected by AI models from contributing humans. “The strongest version of profit sharing is not a tax, but a compensation system based primarily on human contributions that make AI systems valuable,” Fernandez said.

“Them [the AI companies] “A reasonable mechanism would resemble a system of collective governance, similar in spirit to music royalties, rather than calculating the exact value of each ‘token’: AI companies would pay a portion of their model profits into a pool, the total share would be anchored by evidence of how much of the model performance is due to data, and payments would be distributed among creators, publishers, platforms or other intermediaries based on audited metrics of data contribution,” Fernandez said.

But researchers Nicholas Vincent and Brent Hecht, of Simon Fraser University and Northwestern University respectively, caution against this approach. One 2023 survey Examining whether it is possible to adequately compensate individuals for their contributions to an AI system, Vincent and Hecht argue that attaching valuations to each person’s data can be highly subjective and potentially counterintuitive.

“Seemingly minor design choices can dramatically alter the distribution of data values, which is a serious concern for any human-AI system that attempts to combine such values ​​for payments or other purposes,” they say. “If a technology depends on the collective contributions of millions or billions of people, we already know that the value of each individual will be very small, so why spend time and energy on performance? [potentially costly] data value estimation?” they concluded.

Creating strong new unions for the 21st century

But direct payments may not be the only way to achieve a more equitable data sourcing process. Matt Prewitt, president RadicalxChange The foundation and one of the two authors of this policy paper advocate for the creation of a new class of legal rights that give people the power to shape how AI works; this is the 21st century version of unions, where “people cannot sign away these rights at an individual level, but instead people must come together in associations to exercise these rights.”

Prewitt said this would create a new class of regulated associations that “have a very serious seat at the table with AI companies and the power to gain shares, fees, management and power.”

Economist and technologist Glen Weyl, principal researcher at Microsoft and founder of RadicalxChange, argues that the goal is not necessarily government or public ownership of companies. According to RadicalxChange, efforts to “divide and fragment property (i.e., give a share of traditional property to more or different individuals) or consolidate ownership (i.e., place it in the hands of some representatives of the public, such as the state)” may provide some benefit, but are better viewed as “mere band-aids.”

RadicalxChange staff wrote: “Fragmenting ownership ‘spreads out’ the same old exploitative incentives of traditional ownership, while consolidating ownership ‘puts all your eggs in one basket,’ intensifying the risks of corporate capture and illegitimate representation.” a piece of policy New models centered on common ownership are being discussed.

New corporate taxes, fewer working hours

Others say mechanisms already exist to enable policymakers to create a fairer AI economy without resorting to untested ideas. These include stronger corporate taxes, antitrust enforcement and workforce protection, according to Dean Baker, co-founder and economist at the Center for Economic and Policy Research.

Although Baker said he was not convinced that artificial intelligence would massively displace human labor, he added that he would still resort to “old remedies.”

Baker said those remedies could include “an applicable corporate income tax at a higher rate for all corporations.” However, he added that the payment method may be new. “The best way to do this is to require companies to transfer non-voting shares equal to the target tax rate (e.g. 25% of shares for a 25% tax rate),” he said.

Additionally, Baker says, when strictly enforced, antitrust enforcement provides a reasonable path toward more equitable economic distribution of AI profits. Baker made an analogy about cheap Chinese products replacing blue-collar labor. “We have allowed Chinese manufactured goods to screw over large swaths of the blue-collar workforce. We should not indulge in protectionism that will keep Elon Musk and Mark Zuckerberg ridiculously rich,” he said.

A lax attitude towards the rise of social media and failure to repeat past policy mistakes, including early global outsourcing, are high on the radar of some of the most senior policymakers in the US, who believe that AI, and especially its employment dimension, will grow as an electoral and social issue in the coming years.

The idea of ​​a universal basic income — or a “universal high incomeThere is already a global precedent for a simpler labor market mechanism to distribute future economic productivity created by AI, a topic that has been debated for years about what Elon Musk calls a “mass unemployment fight” program.

The answer is not to work at all, but to work less.

“We set the 40-hour work week limit 90 years ago and it hasn’t changed since,” Baker said. “Other countries have shortened the workweek, the working year. If AI is going to give us the promised boom in productivity, let’s lower the threshold to 32 hours, possibly even lower. We can also double the overtime bonus to 100% instead of 50%,” Baker said.