The actual and potential benefits of A.I. notwithstanding, we need to be far more prepared than we are now to manage, productively, the impact on society
It’s hardly a revelation that AI has grown exponentially. While seizing the imagination—and resources–of investors, A.I.’s rapid development has caught the nation off guard. Not only is the nation less than prepared, there seems little concern for the cost, including the environmental impact to meet the pressing need for electricity to fuel it.
We celebrate the advancements A.I. delivers, particularly in the medical sciences, but as promising as they are, the primary public focus remains on safety and regulation, with safety concerns reflecting the fear of being outsmarted and thus requiring some controls and regulation. But, companies resist government intervention at every turn and countries take chances for fear of missing out on the promise—and financial return–that promoters of A.I. hold out to them. Thus, there is no realistic planning at any level of our government to mitigate risks or steer developments in A.I. to maximize the full benefits of these systems.
Change is Hard, Rapid Change is Harder
Preparing society for significant change without knowing what it will be is daunting. Notwithstanding the scope and rapidity of A.I. developments, though, we have been here before.
At the turn of the century, America was ill prepared for the evaporation of labor intensive manufacturing industries and the communities that lost them were hit hard. If a bounce-back was anticipated, created by new industries to help revitalize communities staggering from the loss of their economic base, it didn’t come. Generations passed before recoveries—requiring highly educated workers, often with college degrees– took place.
The cost to communities while transformations took place included high unemployment, high rates of addiction and premature death.
Labor markets adjust over generations, after all, not overnight.
What do human beings do for work, for purpose and meaning in their lives when they are displaced? For many, work is at the center of life, stabilizing, of course, as work provides the means to support one’s self and others. A job confers security and dignity. When that is lost, society is in trouble.
Artificial Intelligence and the Workforce
The astronomical investment, in the trillions already, with VP Vance giving the figure of “$700 billion, give or take, that’s estimated to be spent on A.I. by 2028,” poses questions about what investments are not being made as a result, and, again, and related, about some of the more specific impacts that the rapidly advancing technology might present, including diversion from other investments and investors’ own operations, securing land, chips and power for data centers that can be costly, of course, and time-consuming to get up and running.
Despite the obvious limits of prognostication, policy makers and others need to be at least thinking, seriously, about who will be left behind. How will they be engaged? How do we encourage, indeed, secure public benefit when A.I. is largely private?
Certainly some of that huge investment—and its return–ought to be directed to the critical challenge of maintaining the legitimacy of an economic system where workers create value and are rewarded, where they hold bargaining power in labor markets, and are not subjugated to the vagaries of the market and the sole prerogatives of capitalism.
A.I. and Challenges to Society
Israeli philosopher Yuval Noah Harari, in “Homo Deus: A Brief History of Tomorrow,” says that wealthy elites will ‘upgrade themselves through tech’ and the masses will be left to rot.’ Harari’s statement is meant to provoke, but it is not difficult to imagine a future taking shape that does precisely what he fears. He isn’t alone. Robert W. McChesney warned of A.I. and the digital revolution wiping out numerous categories of jobs, as he observed, in “People Get Ready: The Fight Against a Jobless Economy and a Citizenless Democracy:” “Capitalism as we know it is a very bad fit for the technological revolution to experience.”
And Daron Acemoglu, a Nobel laureate economist at M.I.T., sees a threat in A.I. moving beyond routine tasks to take over highly skilled work, “eroding the value of human experience and leaving people to handle whatever the machines can’t. That could mean an economy in which the owners of A.I. systems capture most of the rewards and the rest of us are left with the scraps.”
It is clear that A.I. has to be steered in the right direction, ensuring safety, promoting responsible innovation that improves the quality of—and perhaps extends—human life. And if it is to maintain political legitimacy, there will be a need for high levels of taxation. The A.I. pioneer, Geoffrey Hinton, in a recent interview, said that if A.I. is to work for everyone, economically, we need “Socialism.”
Elon Musk has given the subject some thought but he focuses on the end result, as he sees it, not on what we might be doing to shape it. He has declared that A.I. will eliminate most jobs and that societies will have to adopt a universal basic income (U.B.I.) to compensate. He hasn’t said anything, though, about where the tax revenue would come from to fund such a program.
Following Hinton’s thought, it can only come from folks like Musk, and Trump, and their billionaire friends, and, of course, the A.I. players paying a lot more in taxes, including a high tax on capital, to support a comprehensive social-insurance system.
The funding piece is but one piece of the picture.
Looking for evidence that serious policy and program discussions relating to the impact of A.I. are taking place, produces little to see, certainly not in Congress to be sure, and, reportedly, not in the executive suites where high placed tech leaders and investors gather to converse about A.I.. Executives only whisper about the obvious impacts on jobs in the hallways of the tech corridors and tech conferences where furious discussions about growth, and profit, take place. The lack of evident concern for society is worrisome.
A.I. schemes developed with an eye to profits generated by eliminating jobs presents a challenge too great to ignore. Fewer jobs in a society without plans for the displaced is a recipe for disaster. And the tech executives just want to whisper about it? While Congress fiddles? And think tanks, anything going on there? Or within major foundations? Or universities? None apparent.
In February 2024, for example, EN:Insights Forum: Assessing A.I.’s Impact on the Workplace offered some insights—shorter work week, “upskilling” within companies—and so did an effort by the deliberative forum at Davidson College. The latter took a look at how individuals, communities and policy-makers should relate to A.I. but limited itself to panel presentations from three critically involved fields, and, notably, with no societal perspective among them.
The limited frameworks within which conversations are taking place, in short, give short shrift to the societal challenges, to the economic and social impacts of a major employer closing or a firm that is functioning without the need it once had for human workers.
We are reminded that a human employee works for eight hours a day but a machine can work for 24 hours, doesn’t need a break and certainly doesn’t negotiate terms and conditions of employment.
Tech start-ups need fewer workers. One company, Gamma, for example, has hired only 28 people and A.I to do what 200 would otherwise be doing. “Tiny teams” are the thing now bringing “bragging rights for making the most revenue with the fewest workers.”
Starbucks, relying on A.I., is cutting jobs, “removing layers and duplication and creating small, more nimble teams.”
We are also seeing reports of reductions in force in such places as banks and engineering firms.
White collar workers are experiencing layoffs as rapid advances in A.I., not to mention Trump’s assault on the civil service, reflects a decline in “knowledge work.” Given the steeper rise in unemployment for college graduates—30% since 2022—more than the 18% overall of other groups–and slower wage growth, college enrollment is taking hits as well. If there is less of a need for employees with college degrees, it is, in part, at least due to the roles that A.I. can fill through automation (e.g. bookkeeping; software coding; accounting; drafting legal documents), given A.I.’s “creativity” and “cognitive skills.”
Creative careers, magazine publishing, newspaper journalism, photography, graphic design, editing, advertising, music, film, T.V., there’s a good chance that people in those occupations are now doing something else for work because those industries have shrunk or transformed themselves radically, shutting out those whose skills were once in high demand.
Developing Alternatives for Displaced Workers
The nation is not known for its capacity to plan for the fallout from change, particularly when it comes to displaced work, shuttered companies, or industries made redundant. It doesn’t take much imagination to project where the U.S. will be as A.I. grows exponentially without concomitant job growth, except, perhaps, as it promises, within its own domains.
Robert Capps, in “People Skills,” in the New York Times special edition magazine (June 22, 2025) “Learning to Live with A.I.” provides valuable perspective. Data, for example, shows that 70% of the skills an average job will require will change by 2030; nine million jobs are expected to be displaced by A.I. and other emergent technologies in that time period; A.I. will create jobs too, by 2030, projecting some 11 million new jobs, for roles that have never existed before. What a challenge for society!
We need some blue-sky thinking and we need it now.
When VP Vance made his first major policy speech since becoming vice president, he framed A.I. as an economic turning point but cautioned that “at this moment, we face the extraordinary prospect of a new industrial revolution, one on par with the invention of the steam engine.” He left it at that, however, focusing on growth and minimal regulation so as not to slow the realization of its potential. Not a word about societal impact.
If the arrival of A.I. is “a new industrial revolution,” we need a coherent reindustrialization strategy to go with it, one that addresses what it means for our society and what to do about it.
G.I. Bill: Adapting the Concept and the Model
The Serviceman’s Readjustment Act of 1944, commonly known as the G.I. Bill of Rights, helped transform America in the postwar era. It was a way to keep the millions of returning veterans from immediately returning to the workforce, because of fears that the Great Depression and mass unemployment would return after the war if too many men looked for jobs after demobilization.
Seeing a parallel, given the nature and scope of the potential displaced workers of our new industrial revolution, it’s worth taking a serious look at the success of this legislation.
The bill included all veterans without regard to class and race, offering support for veterans to go to college, providing tuition, books, fees, and a stipend. Technical training programs were also included as was a low-interest loan provision that allowed veterans to secure a mortgage for new housing construction, acquire a home or a farm with no down payment, which helped speed the development of the postwar suburbs.
By 1956, 7.8 million veterans had used the G.I. Bill education benefits, some 2.2 million to attend colleges or universities and an additional 5.6 million for some kind of training program.
Historians and economists judge the G.I. Bill to have been a major political and economic success, and a major contribution to the nation’s stock of human capital that encouraged long-term economic growth.
As much as any piece of legislation passed in the postwar period, the G.I. Bill helped propel thousands of veterans into the middle class. It set the stage for postwar abundance and prosperity and contributed to the growth of higher education opportunities as a ticket to middle-class status in America.
So, there is a foundational principle here and a model framework for federal legislation that suggests how the nation can prepare for a new age, one that will include major shifts in workforce needs, significant increases in displaced workers, provide opportunities for training for new jobs, for the nation’s growth and sustainability, and, at the same time, plans for social support.
States could certainly offer partial solutions as well, blending training, say, with desirable economic and environmental objectives such as what New Jersey is doing with its $4.3 million program to prepare New Jersey residents for green economy careers while also serving overburdened communities.
Governor Phil Murphy called the program “a unique chance for entities across New Jersey to train residents for jobs in clean energy and other vital green sectors, which will help support a stronger and more diverse economy.”
In addition to federal and state policy and programming, A.I. companies have to assume a significant financial role in the new age they are instrumental in launching. Higher taxes on A.I. profits, as noted, have to be anticipated. Robert A.G. Monks, the prominent entrepreneur and writer who railed against “The Imperial C.E.O.,” held that a company that creates a problem is not exempted from trying to find, and I’d add, helping to fund, a solution to the problem.
Preparing for Transition
The task for us, then, as we continue to support innovative advances on so many fronts–including A.I., positioned as a growth industry that promises jobs with good incomes for people–is to steer its work toward building a stable and equitable society, improving people’s lives; repairing ecosystems, rebuilding infrastructure and expanding housing.
Modern manufacturing requires high technology, requiring different skills from those needed in the 20th century. We can innovate in pharmaceuticals, clean-energy technology, robotics and semiconductors, but those innovations will require “infrastructure” to support its development, education and workforce training, and, frankly, cooperative alliances with global partners.
Any effort to create an alternative future that works, however, will not be without conflict between those who see A.I.’s potential to improve decision-making and productivity while avoiding concern about the impact on society and those who warn of the downside if A.I. ceaselessly eliminates tasks and jobs, over-centralizes information and discourages human inquiry and experiential learning, and, at the same time, empowers a few companies to rule over our lives, reinforcing a two tier society with even greater inequalities and status differences.
In such a scenario, A.I., according to Daron Acemoglu, “may even destroy democracy and human civilization as we know it.’ He fears that “this is the direction in which we are heading.”
The challenge is hardly for the faint-hearted but it is not one we can wisely–or safely–avoid. And, if we can launch solid public-private partnerships, effectively manage conflicts while keeping our objectives in mind, accept the role for government that is essential to the picture, acknowledge the need for significant tax changes, and, no doubt, attract vigorous participation by the so-called third sector, including universities and philanthropies, we can be better prepared for the massive changes that are surely just around the corner.