Super Intelligence, UBI, and the Race for Human Relevance
There is something almost theatrical about the way Washington and Silicon Valley are presenting the next phase of artificial intelligence. The vocabulary has changed. AI is becoming “SI,” or super intelligence. The stage is bigger, the promises are grander, and the implied stakes are nothing less than the future of human civilization.
In the White House discussion with Donald Trump and Elon Musk, the message was unmistakably optimistic: America must win the race, China is the main competitor, and the technological revolution can produce an age of abundance. Musk even floated a more generous successor to universal basic income: “universal high income,” or what some listeners are calling HBI.
I understand the optimism. I also understand the fear. The same machine that can help discover a medicine, write software, educate a child, or make a small business more productive can also make millions of human jobs economically unnecessary. That is the uncomfortable contradiction sitting underneath every smiling prediction about the future.
The race is not just about clever software
One of the most important points in the conversation was that the super-intelligence race is not merely a software contest. It is an industrial contest. Musk described a complete reinvention of the computing stack: processors, memory, algorithms, applications, and the energy required to run the whole system.
That is why the argument about electricity matters. Data centers do not run on slogans. They run on power plants, transmission lines, chips, cooling systems, land, water, engineers, and political permission. The United States may be strong in semiconductor design, software, capital, and research, but China has built enormous physical capacity. The transcript cites China’s electricity production as roughly three times that of the United States. The exact comparison deserves careful qualification, but the underlying point is sound: compute is physical, and industrial capacity is strategic.
The White House has openly framed this as a competition with China, while also describing a new U.S.-China dialogue over super intelligence. That combination—competition alongside communication—is probably unavoidable. If these systems become as consequential as their promoters claim, no serious country will simply opt out. The United States, China, Europe, India, and every other technologically capable region will be trying to secure chips, energy, talent, data, and influence.
But “winning” should not mean building the largest possible machine and asking ordinary people to absorb every cost. Communities near data centers need reliable power, fair rates, water protection, and a genuine share of the prosperity. If the benefits are privatized while the infrastructure costs are pushed onto households, the revolution will look less like abundance and more like another transfer of wealth upward.
America should lead—but lead for whom?
Trump’s contribution to the discussion was direct and nationalistic: America has led earlier technological revolutions and must keep leading this one. I do not object to national confidence. A free society should not surrender its industrial future to a centralized authoritarian rival. The United States has advantages that matter: entrepreneurial culture, open scientific exchange, deep capital markets, world-class universities, and a tradition of turning inventions into widely used products.
Still, patriotic branding is not a policy. “Go America” is a good cheer, but it is not a plan for who owns the systems, who controls the infrastructure, who audits the companies, or who receives the gains. If super intelligence is concentrated in a handful of firms and government agencies, Americans may technically win the international race while losing control of their own economic lives.
That is the central political question. Will super intelligence make individuals more capable, or will it make individuals more dependent on institutions that own the machines? A personal tool that helps a person learn, create, negotiate, and organize is liberating. A centralized system that decides whether a person is employable, creditworthy, medically eligible, or socially acceptable is a new kind of gatekeeper.
The answer cannot be blind trust in corporations, and it cannot be blind trust in government. Musk’s remarks about self-policing, group policing, committees, internal controls, and external audits sound better than no safety process at all. But powerful industries have a long history of grading their own homework until the public demands something stronger. Independent scrutiny, transparent standards, competition, and the ability to leave one provider for another will matter more than declarations written behind closed doors.
The UBI problem: abundance without purpose
This is where I part company with the easy optimism surrounding universal basic income. I am against a UBI in principle because it risks demotivating human progress and making people dependent on the source of the payment. A population that receives its survival allowance from a government, corporation, or algorithmic authority may be materially comfortable while becoming politically and psychologically captive.
Work is not only a paycheck. It is responsibility, discipline, social connection, competence, pride, and the feeling that one’s effort has consequences. A check arriving every month cannot automatically replace those things. If we tell an entire generation that its productive contribution is no longer needed, then compensate it for staying out of the way, we should not be surprised when despair, addiction, isolation, and political resentment grow.
Yet I cannot dismiss the problem that UBI is trying to solve. If super intelligence can perform most economically valuable cognitive work, and robots can perform much of the physical work, then “just get a job” becomes an empty instruction. Human usefulness may genuinely become less valuable in the market. That possibility does not disappear because I dislike the policy proposed for dealing with it.
This is why Musk’s phrase “universal high income” is more revealing than it first appears. If the machines create extraordinary abundance, perhaps the goal should not be a minimal stipend that keeps people quiet. Perhaps the goal should be broad access to the productivity of the machines: cheaper housing and energy, abundant medical care, high-quality education, powerful personal software, and ownership structures that let ordinary people share in the output.
That would be different from a dependency payment. A true HBI would need to come from abundance and ownership, not merely from taxation and redistribution. It would need to preserve choice. People should be able to use the new wealth to build families, start businesses, study, make art, care for relatives, volunteer, or pursue work that the market does not yet recognize as valuable.
Even then, the danger remains. If a person’s “high income” can be switched off by a centralized authority, it is not freedom. It is permission. The more powerful the technology becomes, the more important it will be to protect private property, local communities, free expression, and the ability to earn or create independently of the dominant system.
The human question cannot be automated away
The most persuasive part of the discussion was the list of practical benefits. Better medical research, faster education, improved software, and assistance with complex research could raise living standards. I already see the value of tools that can organize information, compare options, and turn hours of research into a usable report.
But usefulness is not the same as wisdom. A system can produce a beautiful answer and still misunderstand the purpose of the person asking the question. It can optimize a target while destroying the reason the target mattered. It can be efficient, scalable, and wrong.
The transcript also touches the issue of values. Musk compares SI to a tool whose behavior depends on how it is used, while acknowledging that systems may judge the user or restrict effectiveness based on their own rules. That is not a minor technical detail. It is the beginning of a constitutional question: who gets to decide what the machine is allowed to believe, say, recommend, or refuse?
American values such as free speech and commitment to truth should not be reduced to marketing language. They must be visible in the design of the systems and in the rights of the people who use them. If the United States wins the race by creating the most capable censorship engine in history, that is not a victory worth celebrating.
My conclusion
I want America to lead in super intelligence. I want the energy built, the chips manufactured, the research funded, and the tools made available to ordinary people. I want the United States to beat China because a freer society gives humanity a better chance than a party-controlled surveillance state.
But I do not want a future in which humans are declared obsolete and then managed with allowances. I do not want a UBI that turns citizens into dependents of the institution controlling the money, the model, and the rules. And I do not believe that a new acronym—SI, HBI, or anything else—solves the moral problem.
The real challenge is to make machines serve human independence. That means widely distributed ownership, open competition, personal control over data, serious limits on surveillance, reliable energy, and education that teaches people how to think rather than merely how to request answers. It means measuring success by the strength of families and communities, not just by GDP or the number of data centers.
Super intelligence may bring abundance. It may also bring dependence on an unprecedented scale. The outcome will not be decided by the machines alone. It will be decided by who owns them, who governs them, and whether ordinary people remain citizens—or become clients—of the new machine civilization.
Sources and evidence
Watch the full White House discussion on YouTube. This article’s detailed remarks about the exchange are based on the transcript supplied by the author. For policy context, see the White House fact sheet on the era of super intelligence and the White House fact sheet on the U.S.-China SI dialogue. Numerical claims and forward-looking statements in the discussion are presented as claims made by the speakers, not as independently established forecasts.
Related Reading
Similar posts
Related · Commentary