Robot Tax: Sanders' Crusade Between Ethics and Pragmatism. Taxing Innovation?
When the Senate HELP Committee published its report on October 6, 2025, the figure was staggering: one hundred million American jobs could disappear in the next decade, wiped out by artificial intelligence and automation. The paradox is deliciously meta: to arrive at this apocalyptic prediction, Democratic staffers sought help from none other than ChatGPT, consulting the digital oracle to predict their own professional obsolescence. It's like asking the Terminator for advice on home security.
The numbers emerging from the study paint a picture of a cyberpunk dystopia: 89% of fast-food workers destined for unemployment, 83% of customer service representatives replaceable, 64% of accountants redundant, and even 54% of software developers who could be replaced by the very technologies they help create. It's as if the engineers of the Death Star designed their own weak point without realizing it, only this time there's no Luke Skywalker on the horizon, but Bernie Sanders with a legislative proposal instead of a lightsaber.
The Vermont senator, Ranking Member of the Committee, did not mince words: "The same handful of oligarchs who have rigged our economy for decades—Elon Musk, Larry Ellison, Mark Zuckerberg, Jeff Bezos, and others—are now racing as fast as they can to replace human workers with what they call 'artificial labor'." The language is that of class warfare, updated for the digital age, where the robber barons don't build railroads but train language models.
Bernie Sanders: The Crusader of the Digital Age
To understand Sanders' proposal on the robot tax, one must understand the man behind the thick glasses and perpetually disheveled hair. Bernie Sanders is not a newcomer to the battle against corporate giants. His democratic socialism—which in America sounds almost like an oxymoron—has roots that go back to the 1960s, when he demonstrated for civil rights and against the Vietnam War. His ideological consistency is proverbial, to the point that his opponents consider him stubborn and his supporters see him as incorruptible.
In the 1990s, he fiercely opposed NAFTA and free trade agreements, correctly predicting that they would hollow out the American industrial Midwest. He fought against Amazon until he forced the Bezos behemoth to raise the minimum wage to $15 an hour. He targeted Walmart for its workers' conditions. He denounced pharmaceutical companies for inflated drug prices. In short, if there were a Frequent Fighter Program for politicians who challenge mega-corporations, Sanders would have accumulated enough miles to go to Alpha Centauri and back.
His position on artificial intelligence is perfectly consistent with this narrative: technology is not neutral, but reflects the choices of those who control it. And if those who control it are the billionaires of Silicon Valley, then the benefits will inevitably flow upwards, while the social costs will be distributed downwards. It is the trickle-down economics theory reversed: instead of wealth trickling down from the top, we have unemployment cascading down on the middle and lower classes.
The Proposal: Taxing the Future
But what exactly does Sanders propose? The robot tax is the centerpiece of a broader legislative package he is preparing, although according to the Washington Examiner, the technical details have yet to be defined. The basic idea is simple in its radicalism: companies that replace human workers with artificial intelligence systems or robots should pay a specific tax. The funds raised would go to finance job retraining programs, subsidies for unemployed workers, and investments in education.
The exact mechanism remains to be defined—and here the technical problems that we will analyze later begin—but the intent is clear: to make companies pay the social costs of automation. Currently, when Amazon installs robots in its warehouses and lays off thousands of workers, the profits remain with Amazon while the social costs (unemployment benefits, healthcare, community degradation) fall on the community. The robot tax wants to internalize these negative externalities, to use economic jargon.
The broader package includes proposals that sound almost utopian in the American context: a 32-hour work week with no loss of pay, the requirement for large corporations to distribute at least 20% of their shares to employees, worker participation on boards of directors at 45% as is the case in Germany, and a ban on companies buying back their own shares—a practice that artificially inflates stock prices and enriches shareholders without creating real value.
Sanders wrote: "We need a world where people live healthier, happier, and more fulfilling lives—not one where machines make all the money." It is a vision that resonates with millions of Americans who see their standard of living eroding despite continued productivity growth, a paradox that has defined the American economy for at least forty years.
Bill Gates and the Divided Tycoons
If there is one surprising aspect to this story, it is that Sanders is not completely alone among the billionaires. In 2017, Bill Gates publicly proposed a robot tax, arguing that governments should slow the adoption of automation in areas where it directly replaces human workers. "If a human worker earns $50,000 working in a factory, that income is taxed," Gates argued. "If a robot does the same job, we should tax the robot at a similar level."
Gates' position is particularly interesting because it comes from within the tech ecosystem. He is not a Luddite who wants to destroy mechanical looms, but one of the fathers of the computer revolution who suggests pumping the brakes. His proposal was not aimed at stopping technological progress but at slowing it down enough to allow society to adapt, using the funds raised to retrain workers and finance sectors where human intervention remains irreplaceable, such as elderly care and education.
Mark Cuban, the billionaire entrepreneur and television personality, has also expressed openness to forms of taxation that redistribute the benefits of automation. Elizabeth Warren, a Democratic senator from Massachusetts, has supported positions similar to Sanders', placing the issue in the broader debate on taxing large tech corporations.
On the other side of the spectrum, Elon Musk has a radically different vision. He has prophesied that by 2040 there will be more humanoid robots than humans, and that "probably none of us will have a job." His solution? A universal basic income, because work will become optional, "like a hobby." It is a vision that oscillates between the post-scarcity utopia of Star Trek and the corporate dystopia of Ready Player One, where everyone lives on government subsidies while immersing themselves in virtual realities to escape the misery of the real world.
Jeff Bezos, who has more to lose than anyone else considering how much Amazon has invested in warehouse automation and robotic logistics, has maintained a strategic silence on the subject. His empire is built on algorithmic efficiency and the systematic replacement of human labor with automated systems, so any public stance would be politically costly.
The Republican Front: Between Innovation and Populism
If you were expecting a united Republican front against the robot tax, prepare for a surprise. The party is experiencing an interesting split between traditional libertarians and new-generation conservative populists, and the issue of artificial intelligence is sharpening this division.
Senator Lindsey Graham dismissed the proposal as "dead on arrival," using the classic Republican rhetoric that any regulation stifles innovation and harms American competitiveness. Ron Johnson evoked the specter of historical productivity: every technological revolution has created more jobs than it has destroyed, so why should this one be different?
This is the strongest argument against the robot tax, supported by impressive historical data. During the Industrial Revolution, despite agricultural automation, unemployment did not explode but transformed, with millions of workers moving from farms to factories. The computerization of the 1980s and 1990s did not produce mass unemployment but created entire new industries. Even in the American auto industry, according to data cited by the conservative think tank AEI, between 2010 and 2022, about 60,000 industrial robots were installed, but 230,000 new human jobs were created.
But there is a crack in this front. Senator Josh Hawley of Missouri represents a new generation of conservative populists who have serious concerns about the impact of AI on workers. In a speech at the National Conservatism Conference in September 2025, Hawley painted a bleak picture: "The farmer, the assembly-line worker, the construction worker with a hard hat and a hammer: all these men live by their bodies, by their labor. There is no place for them in the transhumanist utopia, where everything is silicon."
Hawley has introduced several legislative proposals on AI, including the AI LEAD Act, which would require companies using automation to invest in worker retraining. He also co-sponsored with Democrat Richard Durbin a bill that would make AI companies liable under product safety laws. In an interview with Axios in July, he stated: "If the Republican Party wants to be the party of working people, we should start thinking about how AI will affect ordinary people."
This split is significant because it reflects a deeper tension in contemporary American conservatism between traditional pro-business ideology and a new economic nationalism that puts American workers at the center. Hawley has even supported raising the federal minimum wage to $15, a position that just a few years ago would have been anathema to any Republican.
The Scenarios: Between Dystopia and Opportunity
Now we get to the heart of the matter: what would really happen if a robot tax were implemented? And what will happen if it is not? The possible scenarios are multiple and the consequences are far from certain.
The dystopian scenario is the one painted by Sanders and increasingly shared even by some repentant technologists: an extreme social polarization where a tiny elite controls the means of automated production and accumulates unimaginable wealth, while growing masses of people lose not only their income but also the sense of identity and purpose that comes from work. Amazon is the emblematic example: in 2023, it laid off 27,000 employees while recording profits of $59.2 billion. UnitedHealth cut staff while its executives pocketed compensation in the tens of millions. Foxconn, the Taiwanese electronics giant, has announced plans to replace 30% of its workforce with robots by 2025.
Sanders' report cites concrete cases of companies openly telling investors that AI will allow them to "cut personnel costs." Salesforce advertises "digital labor" platforms with the slogan "stop hiring humans." Autonomous truck startups boast that their technology eliminates the "problem" of higher wages for drivers. It is disturbing how explicit the contempt for human capital is.
The opposite scenario, painted by critics of the robot tax, foresees a slowdown in American technological innovation just as China and other countries accelerate. If American companies have to pay extra taxes to automate, they argue, they will simply move production abroad or be overtaken by international competitors who do not have these constraints. The paradoxical result would be a net loss of competitiveness without really protecting workers, because their jobs would be lost anyway to foreign companies.
The libertarian think tank Reason compared the proposal to "taxing the Model T in the early twentieth century." Would it have saved carriage makers and blacksmiths? Perhaps temporarily. But it would also have slowed down a technological revolution that ultimately greatly increased overall prosperity and created millions of new types of jobs that no one in 1910 could have imagined.
Then there is an intermediate, more nuanced scenario. AI might not replace workers entirely but radically transform the nature of work. Instead of 100 million unemployed, we could have 100 million workers doing very different things than today, perhaps earning more because AI has eliminated the repetitive and boring parts of their jobs, allowing them to focus on more creative and high-value aspects. It is the optimistic "AI as a copilot" scenario, where technology amplifies human capabilities instead of replacing them.
But there is a problem with this rosy scenario: the speed of change. Previous technological revolutions took place over decades or centuries, allowing societies to adapt gradually. The Industrial Revolution took more than a century. The agricultural revolution, thousands of years. Artificial intelligence is compressing similar transformations into less than a decade. As the HELP Committee report notes, this unprecedented speed may not leave enough time for retraining and adaptation.
The Shadows of the Proposal
Even if one wanted to implement a robot tax, the technical and practical challenges are formidable. The first and most obvious: how do you define a taxable "robot" or AI system? Does software that automates accounting count? What about an algorithm that optimizes delivery routes? If Amazon uses AI to predict demand and reduce planning staff, how do you calculate how many equivalent "robots" it has employed?
This is not a theoretical concern. In 2017, the European Parliament voted against a robot tax proposal precisely because technicians could not find a sensible operational definition of what to tax. The risk is creating legislation so vague as to be unenforceable or so specific as to be easily circumvented with legal subterfuge.
Then there is the issue of international competitiveness. If the United States unilaterally implements a robot tax, companies could simply move their operations to more welcoming countries. This is particularly problematic for the technology industry, where software can be developed practically anywhere in the world. Amazon could keep its robotic warehouses in America but move all its software development to Bangalore or Warsaw.
More orthodox economists raise a more fundamental objection: taxing technological innovation is always a bad idea because in the long run it reduces productivity and impoverishes everyone. If in 1920 we had taxed tractors to protect manual farm workers, today we would have more expensive food and a poorer population. The right solution, they argue, is not to slow down progress but to redistribute its benefits through progressive tax systems on profits and income, not on the technology itself.
There is also an element of distributive justice that complicates the picture. The American Enterprise Institute criticized Sanders' report for ignoring data showing that AI is actually reducing some forms of inequality by allowing less-skilled workers to be more productive. An AI virtual assistant allows a small business to compete with a large corporation without hiring expensive staff. An automatic translator allows an Italian freelancer to work for American clients. Taxing these tools could paradoxically harm the very workers the tax is intended to protect.
The Human Aspect
Behind the statistics and economic projections are real people with real lives. When a fifty-year-old truck driver from Kansas loses his job to an autonomous driving system, it's not just an entry in a spreadsheet. It's a man who may have done that job for thirty years, who has a mortgage to pay and children to send to college, who identifies deeply with his profession. "I'm a truck driver" is not just a job description, it's an identity.
Sanders' report emphasizes these human aspects, perhaps because Sanders himself has always had a talent for making economic abstractions concrete. When he talks about the 3 million fast-food workers who could lose their jobs—that 89% cited in the study—they are not numbers but people working grueling shifts for minimum wage, often without health insurance, who see even that fragile economic security evaporate.
Software developers, that 54% at risk, present a particularly ironic and tragic case. They are the people who built the technology industry, who wrote the code that now threatens their own jobs. Many have accumulated huge student debts to graduate in computer science, attracted by promises of lucrative and secure careers. Discovering that generative AI can write decent code in a fraction of the time it would take a human must sound like a cosmic betrayal.
There is a profound injustice in the fact that the benefits of automation are asymmetrical. When a company automates and lays off workers, the additional profits go almost entirely to shareholders and top managers. Jeff Bezos gets $50 billion richer, while the warehouse worker who lost his job struggles to pay rent. It is not only economically inefficient—all that lost aggregate demand harms the economy as a whole—but it is morally unacceptable to many.
Yet there is also the other side of the coin. Many jobs that AI could eliminate are objectively terrible. No one dreams of spending eight hours a day in front of a deep fryer for minimum wage. No one aspires to spend their life in an Amazon warehouse scanning packages under algorithmic surveillance that measures every second of break time. If automation truly freed people from these alienating jobs, allowing them to do something more meaningful and well-paid, it would be a net gain for humanity.
The problem is that "if." If there is adequate retraining. If there are new, better jobs available. If society can adapt to the speed of change. It's a very big "if," and so far recent history is not encouraging. The manufacturing workers in the Midwest who lost their jobs to globalization did not magically transform into programmers. Many ended up in low-wage service jobs or left the workforce entirely, fueling opioid crises and economic despair that have reshaped American politics.
Towards Which Future?
We come to the heart of the matter: is Sanders' proposal a forward-thinking vision that could save the American middle class, or is it a protectionist illusion destined to fail against the unstoppable forces of technological progress?
The truth is probably somewhere in the middle, and much more complicated than any political slogan can capture. Artificial intelligence and automation will advance, that is inevitable. The question is not "if" but "how" and "for whose benefit." Sanders is absolutely right to point out that without political intervention, the default is that benefits flow upwards and costs downwards. Markets do not spontaneously solve this problem.
At the same time, a rigid and poorly implemented robot tax could be counterproductive. The definitional challenges are real, the risks of capital flight are real, the danger of stifling innovation is real. Perhaps the answer is not a specific tax on technology but a broader rethinking of the tax system, where profits and wealth are taxed more aggressively while the burden on wages is lightened.
Alternatives to the robot tax include universal basic income, supported by figures as diverse as Andrew Yang and, unexpectedly, Elon Musk himself. The idea is that in an increasingly automated economy, the link between work and income must be broken. Every citizen would receive an unconditional monthly payment that guarantees a minimum level of subsistence, financed by the profits generated by the automated economy. It is radical but has its own logic: if robots do all the work, why should we still measure human value through employment?
Another path is to invest massively in education and retraining, not the symbolic programs of today but something much more ambitious. Singapore has implemented the SkillsFuture program, which guarantees lifelong continuing education credits to all citizens. Denmark has a "flexicurity" system that makes it easy for companies to lay off workers but provides generous subsidies and retraining for workers. These models could be adapted to America, although they would require a public investment on a scale that currently seems politically impossible.
There is also the possibility of directly regulating how AI is implemented in the workplace, instead of taxing it. Mandatory transition periods, company-funded retraining packages, or worker representation in automation decisions could be required. This is the approach favored by unions and has the advantage of directly addressing the problem of power in the employment relationship.
The crucial gap is between the speed of corporate innovation and the slowness of the legislative response. Tech companies move at the speed of the internet, legislatures at the speed of parliamentary committees. When Amazon implements a new warehouse automation system, it does so globally in a few months. When Congress debates a legislative response, it can take years before anything is passed, and more years before it is implemented. In the meantime, accomplished facts accumulate.
Sanders at least has the merit of forcefully raising the issue, forcing a public debate on a topic that would otherwise remain confined to the boardrooms of Silicon Valley and the offices of CFOs. His robot tax may be imperfect, it may even be impractical in its current form, but it serves as a Schelling point for a necessary conversation: what kind of technological future do we want? And above all, for whom?
Your Voice
And you, what do you think? Are you on Sanders' side, convinced that companies should pay for the social costs of automation? Or do you believe that taxing innovation is always wrong and that the solution lies elsewhere—in a universal income, in massive investments in education, or simply in letting the market adjust naturally as it has always done in previous technological revolutions?
Have you ever directly experienced the impact of automation or AI in your work? Do you feel threatened by the prospect of being replaced by an algorithm, or do you see AI as a tool that can make you more productive and free you from the most boring parts of your job?
And if tomorrow you were to decide the country's technology policy, what balance would you seek between technological progress and worker protection? Is there a way to have both, or must we necessarily choose?
The question is not academic. The decisions we make today—or postpone—will shape the world of work for decades to come. Bernie Sanders has thrown a stone into the Silicon Valley pond, and the ripples are spreading. Now it is up to all of us to decide how to navigate them.