Bernie Sanders and Donald Trump agree on almost nothing. But they do agree that the public should own a piece of the AI industry.

The Vermont senator and the president disagree on both the structure and stake of public ownership, but the idea is being discussed at the highest levels of government. Even OpenAI and Anthropic back versions of the idea, though Anthropic’s is a tax rather than a stake. Let’s tune in.

The table below summarizes preliminary proposals and shows how far apart they stand, from a voluntary sliver to an outright seizure. After taking a stake in Intel, the president said he wanted “many more cases like it.” Treasury paid $8.9 billion for 9.9% of Intel in August 2025; by the following spring the stake was worth roughly $36 billion, increasing the appetite for such deals. The Pentagon has already taken 15% of a rare-earth miner. This is a pattern, not a one-off.

The argument for these proposals is a public-finance argument, and a strong one. The science under AI grew out of decades of federally funded research. The training data came from the writing, code, and art of millions of people who were never asked and never paid.

Sanders puts the principle plainly: “When a public resource generates wealth, the public should share in that wealth.”

The cleanest versions cost the taxpayer nothing up front, because the equity is contributed rather than bought. That is not the Intel model, which Washington bought for cash; it is the AI version now on the table, where the shares would be donated. If the bubble bursts, the public is out nothing. If it holds, the public owns a slice. A bet with no ante is a rare thing in public finance.

Source	Stake	Structure
Bernie Sanders	Roughly 50% government position (reported figures vary)	Federal sovereign wealth fund; government holds voting shares; ~$1,000-per-person dividend
Trump administration	Case-by-case equity stakes; 9.9% of Intel (now ~$36B)	Direct federal ownership; framed as a taxpayer “windfall”
OpenAI	~5% of equity (~$42.6B) contributed voluntarily	“Public Wealth Fund” modeled on Alaska’s; returns distributed to citizens
Anthropic	No equity	Taxes on AI firms to fund worker support, possibly UBI

Proposals as of July 2026; talks remain preliminary and any federal version would require an act of Congress.

There’s a real danger, though, in what the government becomes when it owns a piece of the industry it is supposed to regulate. A public stake in AI can be a dividend or a trap, and the whole difference lives in the fine print.

Three things separate the dividend from the trap. The first is the size of the stake. The second is a wall between the government as owner and the government as referee, so the hand that banks the dividend never writes the safety rules. The third is a fence around the money: proceeds earmarked for the workers the technology displaces, not swept into the general fund. None of the three enforces itself.

Here’s a loose historical precedent. In 1998, 46 states settled with the tobacco industry for about $206 billion, paid out over 25 years. The states came to lean on the yearly checks, which quietly made them partners in the survival of the product they were supposed to fight. And the money drifted: today states spend only about three cents of every tobacco dollar on the anti-smoking programs the settlement was meant to fund.

A stake with no end date makes the government a permanent co-owner of the industry it regulates, and permanence is one thing that turned a tobacco settlement into a tobacco dependency. The answer is a fixed end date. The same law that creates the stake should set the year it must end. This is known as a sunset clause.

If Uncle Sam owns a stake, he should collect the dividend through the buildout years, then sell it down on a fixed, published schedule until the position is gone. Ten or 15 years. Economists can pick the number. The deadline should be set in law from the start, so a future Congress cannot quietly extend it.

Temporary co-ownership lets the public bank the upside of the boom without leaving the referee holding shares in the game for good. Sanders and Trump, from opposite ends of the political spectrum, have seized on a real grievance and reached for the permanent version of the remedy, which is the version most likely to curdle. Of course, sunset clauses are not etched in stone either.

Another challenge is that the moment Washington owns pieces of its AI champions, other capitals follow — Beijing, Brussels, the Gulf — each taking a stake in its own, and the claim that American platforms answer to no government gets harder to make. A vendor with the state on its cap table is not a neutral one. No wall and no expiration date solves this problem.

A stake also puts the government in the business of picking winners. Own a piece of OpenAI or Anthropic and Washington acquires a financial interest in their business, and a reason to favor them when it writes the next rule or signs the next contract. The startup is forced to compete against incumbents favored by the feds. And in the fast-moving AI field, the players change rapidly.

AI’s economic challenges are real and the grievance underneath these proposals is legitimate, but government ownership is the wrong remedy. The conflict of interest is real, the precedents are bad, and it’s hard to imagine that a referee with money on the game will be neutral.

Still, the momentum is real, too. Sanders, Trump, and the labs are all pushing versions of the same idea, and one of them may pass. If it does, the temporary version with guardrails beats the permanent one: price it honestly, wall it off, aim the money at the damage, give it a hard end date. None of that is a reason to take the stake. It is only what keeps a bad idea from calcifying into a worse one.

Attitudes towards AI differ by country, gender, profession, age, and political affiliation.  A few of those gaps are startling. This article is chock-full of stats. Read it for the surprises, or glance at the bar graph below for a quick overview.

Let’s start with geography, the widest split of all. Ask people in China whether they trust AI and, Edelman finds, nearly nine in 10 say yes; ask Americans and barely a third do. The same chasm shows up, in the Stanford AI Index, on the larger question of whether AI’s benefits outweigh its drawbacks, where most Chinese say it’s good stuff and most Americans have their doubts. 

Here’s a possible explanation. Where economies are young and growing fast, AI reads as a ladder up; where they are mature, it reads as a threat to jobs and more. Trust in AI seems to track two things, confidence in institutions and the expectation of personal gain, and both run higher in many Asian countries than in a wary West.

In the U.S., men are about twice as likely as women to expect AI to be good for society, Pew finds, and the gap is wider still among the researchers who build it. The tempting explanation, that women use the tools less, no longer holds: over the past two years women have drawn even with men in using chatbots, yet they trust them less. Women are also likelier to say AI is moving too fast. 

Adults under 50 reach for ChatGPT at twice the rate of their elders, Pew reports, yet it is the under-30s who are most convinced it will be bad for society. Here, familiarity breeds unease, and for a concrete reason: the young are not only the heaviest users but the most exposed. AI may be coming first for the entry-level jobs they are trying to land, and they sense it, with Gen Z likelier than any older group to expect it to cut into their job prospects, per the Harris Poll. 

Among the AI researchers surveyed, most expect the technology to help the country over the next two decades, Pew’s survey shows; among the public, fewer than one in five do. Some of that is knowledge, since the experts grasp what the systems can and cannot do and fear the lurid scenarios less.

Of course, the people who design AI have their careers and fortunes riding on its success, while the people who answer phones or drive trucks see mainly the threat to their own. The same pattern runs across industries, from technology workers who welcome AI on the job to transportation workers who oppose it. As per Miles’ Law, where you stand depends on where you sit.

The last divide is one that’s moved in recent years, and it’s moved fast. Two years ago Republicans were the AI skeptics; Democrats have since caught up and passed them. Today, just over half of Republicans now trust Washington to regulate AI; barely a third of Democrats do, Pew finds. 

AI companies are now more admired on the right than the left, a Harris Poll shows. Democrats are cooling on companies they once cheered, and Republicans are warming to a boom their side now champions. That said, in both parties more people worry that regulation will do too little than too much; what they split on is whom they trust to do the reining.

Despite some loud voices, there is no single verdict on AI.  Optimism comes from those with the most to gain, in the rising economies and inside the labs; doubts rise from those with the most to lose or the most to fear. Whatever AI turns out to be, it is being built by the people most enthusiastic about it, for a public that is not.

America just turned 250. The founders designed self-government for a world of pamphlets and town meetings, and we now run their political architecture on AI.

The birthday question is whether AI bolsters democracy or undercuts it. Serious thinkers have lined up on both sides with substantial arguments.

Here is my scorecard, distilled from five books and seven articles, and then the question neither side asks: which is growing faster, power over AI or access to it?

Start with surveillance.

Yuval Noah Harari argues in Nexus that a democracy is a distributed information network with self-correcting mechanisms: a free press, opposition parties, and courts that catch mistakes and fix them. A dictatorship is a centralized network that suppresses correction. For two centuries, centralization carried a built-in cost, because total surveillance required armies of human informants, and armies are expensive. AI removes the cost. It watches everyone, all the time, for pennies. The evidence is no longer hypothetical. A study in the Quarterly Journal of Economics documented the feedback loop in China: local unrest leads to government purchases of facial-recognition AI, and those purchases suppress subsequent unrest. The authors titled their paper “AI-tocracy.”

The second argument is economic.

Past technologies replaced particular workers, the switchboard operator, the toll collector, while creating jobs for the people who ran the new machines. AI’s ambition targets the entire workforce. Daron Acemoglu and Simon Johnson devoted a book, Power and Progress, to this worry, writing that “the current path of AI is neither good for the economy nor for democracy.” Acemoglu, a 2024 Nobel laureate, sharpened the point in Fortune this February, warning that on the current path of job destruction and rising inequality, “U.S. democracy is not going to survive.”

The third argument targets the machinery of self-government itself.

I sounded this alarm in Harvard Business Review back in 2019, warning that AI was poised to make high-fidelity forgery of video, audio, and documents cheap and automated, with potentially disastrous consequences for democracy. Forgery is ancient. AI industrializes it. Security technologist Bruce Schneier predicts that AI will optimize lobbying and draft “micro-legislation,” tiny provisions that quietly benefit one group, and he observes that the technology mostly makes the powerful more powerful. He and Nathan Sanders began worrying in earnest when an AI-written letter opposing AI regulation ran in the New York Times. Marietje Schaake supplies the institutional capstone in The Tech Coup: unelected companies now perform functions that once belonged to governments.

The prosecution rests. Now comes the defense.

On July 4, computer scientist Daphne Koller marked the country’s 250th birthday, and her own 37th anniversary as an immigrant, with a visit to Shasta Dam. In a reflection posted that day, she argued that America’s signature achievement is taking what was scarce and making it abundant: water into power at Shasta, electricity into a grid anyone could plug into, computation into a pocket. She has done it herself; Coursera, which she co-founded, put an elite education in front of more than 150 million learners. AI, she wrote, is the next chapter, “making abundant one of the world’s scarcest resources: powerful reasoning.” The judgment once reserved for credentialed specialists now belongs to anyone who can frame the right question. Lawyers and doctors bill by the hour. AI answers by the second.

The economic counter comes from Acemoglu’s MIT colleague David Autor, who argues in Noema that AI can extend expertise to workers without elite credentials and thereby rebuild the hollowed-out middle of the labor market. Early evidence points his way. When a Fortune 500 firm gave its customer-support agents an AI assistant, productivity rose 15% on average, and the gains went overwhelmingly to the newest and least skilled workers, who improved in both speed and quality. The study, published in the Quarterly Journal of Economics, found that the most experienced agents gained little. If the pattern holds, AI could compress the very gaps Acemoglu fears it will widen.

Reid Hoffman and Greg Beato’s Superagency states the optimistic case in general form: AI amplifies individual agency so broadly that the real danger lies in democracies ceding its development to less benevolent actors. In Plurality, Taiwan’s first digital minister Audrey Tang and economist Glen Weyl describe a decade of digital tools that found consensus across a polarized public on live legislation, from ride-sharing rules to pandemic policy. A controlled experiment backs them up. Google DeepMind researchers built an AI mediator, tested it on 5,734 Britons deliberating questions like Brexit and immigration, and reported in Science that participants preferred the AI’s group statements to a human mediator’s, rating them clearer and less biased. The groups also ended up less divided. A town hall has never fit a million people. It might now.

I set the two columns side by side and noticed something odd: they never meet. The pessimists are arguing about who controls AI. The optimists are arguing about who gets to use it. Power and access are different questions, and both camps can be right at the same time.

Koller’s dam makes the point physically. Generation is concentrated, a handful of turbines owned by a few. The grid is distributed, and anyone can plug in. One machine does both at once. AI shares that anatomy: anyone can plug into a frontier model for $20 a month, while the frontier weights and the data centers that train them belong to a half-dozen companies.

Gutenberg adds the time dimension. The press broke Rome’s monopoly on scripture, and four centuries later it built Hearst’s empire; access and power traded places on the same machine. Both forces are real. The open question is which one moves faster, and the current fights over open weights, chip exports, and model ownership are fights that will help settle this question.

The founders faced a similar question about concentrated power and answered it by distributing the vote, narrowly at first, and later to nearly everyone. Koller ended her post with an obligation that fits the country’s 250th year: anyone given more than their share owes the work of making sure the next scarce thing does not stay scarce for long. Intelligence is the next scarce thing. Koller’s dam is already built, along with the frontier models and the data centers that train them. The choice in front of us is whether we also build the grid, providing broad, cheap access to AI for all Americans.

On Tuesday in Shenzhen, the Chinese company UBTech unveiled the U1, a full-sized humanoid robot with silicone skin, blinking lashes, manicured nails, and an AI tuned to read your mood. It comes in male and female versions, and racked up more than 13,000 orders by the end of launch day, with deliveries beginning in September.

“It will never betray you, will always be loyal to you, and will love you unconditionally,” promised Michael Tam, the executive running UBTech’s consumer brand.

The sci-fi TV series “Humans” imagined lifelike android “synths” sold to ordinary families as helpers and companions, and it treated the idea as speculative fiction. A decade later, the fiction has a September ship date. What it does not have is an American logo.

Elon Musk announced the Tesla Bot in 2021 and has been re-announcing it ever since. He hoped for production readiness by 2023. Entering 2025 he targeted 10,000 units, then trimmed the goal to 5,000.

The unveiling of Optimus 3, promised for March of this year, slipped because the robot needed “finishing touches,” and as of Tesla’s April earnings call Optimus 3 is still MIA, with the reveal now promised for late July or August. Tesla is spending $20 billion in capital expenditure this year, with Fremont assembly lines converting from the Model S to Optimus. The robot is not vaporware; it’s merely years behind schedule.

Now look at what China shipped while Optimus was getting its finishing touches.

In April, a bright-red humanoid named Lightning, built by smartphone maker Honor, ran Beijing’s E-Town half marathon in 50 minutes and 26 seconds, roughly seven minutes faster than the human world record. The remarkable number is not the 50 minutes. It is the comparison to last year’s inaugural race, when the winning robot needed 2 hours and 40 minutes and most of the field fell over, wandered off course, or lay down at the starting line. The machines cut their time by two-thirds in 12 months.

Meanwhile, UBTech won a $37 million contract to deploy its Walker S2 humanoids at the Fangchenggang border crossing with Vietnam, where they guide travelers, patrol corridors, and inspect cargo. Barclays estimates China accounted for 85% of the world’s humanoid robot installations last year, and Beijing counts more than 140 domestic companies selling over 330 models.

Why the gap? Talent is not the problem, and neither is money. The difference is the customer.

Optimus’s most important customer has always been the Tesla shareholder, and a Musk keynote serves that customer just fine. The Walker S2’s customer is a border authority with a delivery date and a cargo queue that does not pause for a reboot.

China’s supply chain proximity and its government’s decision to treat humanoids as a strategic industry help, but the deeper difference is that Chinese robot makers get paid for delivery while Optimus gets valued for anticipation. Only one of these incentive structures produces robots in a timely manner.

In fairness, the most useful robots in American homes and hospitals are not humanoid. Form follows task, and when the task is specific, the human form is expensive overhead. For instance, the da Vinci surgical system, which has operated on more than 20 million patients, is four arms bolted to a cart, because a surgeon needs wrists steadier than human wrists and has no use for a reassuring face. The most successful household robot in history is a disc that eats dust. No one wants their Roomba to watch the sunset with them. 

The humanoid shape is a bet on generality, on a machine that can use our doorways, our staircases, and our tools. That bet makes sense at a border crossing built for human bodies. It is far less obvious in the operating room.

Companionship has never required human form; ask anyone with a dog. The New York Times recently told the story of Jan Worrell, an 85-year-old widow on a remote stretch of the Washington coast, and her companion robot ElliQ, which resembles a small reading lamp. It has no face, no legs, and no silicone anything, yet it shares her morning coffee, nudges her toward chair yoga, and has become, in her words, “me and my robot.”

Hundreds of ElliQ units deployed through New York State’s Office for the Aging show the same pattern of daily attachment. A machine does not need a body to keep you company, and the ElliQ price tag is much lower.  (Full disclosure: I serve on the board of Intuition Robotics, the maker of ElliQ.)

So why did UBTech give the U1 lifelike skin, styled hair, and a face you can customize to resemble anyone you choose?

Every new medium in memory has been pulled toward intimacy by its early adopters: the VCR conquered the living room on the strength of what people watched in private; the early internet monetized romance and its rougher cousins before it monetized much else; and app stores learned that “companionship” is a category with remarkable elasticity.

A humanoid robot with a skin warm to the touch is heading in a certain direction, whatever its maker’s official positioning. The company states that the U1’s skills don’t extend to the bedroom, then adds “for now.”