Wall Street Jobs Are Falling to AI as Profits Prove the Unfortunate Reality

There is a peculiar kind of candor that only arrives after the quarterly earnings report. Less than four months before Bank of America posted $8.6 billion in first-quarter profit — $1.6 billion more than the same period a year prior — CEO Brian Moynihan was reassuring his 210,000 employees on national television that artificial intelligence was not a threat to their livelihoods. Then the numbers came in, and the reassurances quietly gave way to something closer to the truth.

The bank’s improved bottom line, Moynihan acknowledged in the earnings call, was aided by shedding 1,000 positions through attrition — achieved by, in his own repeated words, “eliminating work and applying technology,” the technology being artificial intelligence. He then predicted more of the same in the months and years ahead. It is a pattern now familiar across the financial sector, where corporate language about AI “augmenting” human workers has been steadily replaced by the language of headcount reduction and operational efficiency.

The numbers tell the story plainly. JPMorgan Chase, Citi, Bank of America, Goldman Sachs, Morgan Stanley, and Wells Fargo collectively recorded $47 billion in profits — up 18 percent — while shedding 15,000 employees. This is not a warning sign on the horizon. It is the horizon.

The Gap Between the Press Release and the Earnings Call

Wall Street has long maintained a carefully curated public posture on artificial intelligence — that it is a tool of empowerment, a force multiplier for human talent rather than a replacement for it. That posture is now visibly crumbling under the weight of its own quarterly data. The more honest framing is this: AI is extraordinarily good at eliminating costs, and in a publicly traded company, eliminating costs is a fiduciary virtue. The workers whose jobs constitute those costs occupy a secondary concern.

At Wells Fargo, the transformation is already granular and operational. AI software is generating instant memos on the creditworthiness of potential borrowers, producing the pitchbooks that banks use to persuade companies to consider merger deals, and rerouting or automatically answering all types of phone calls from credit card customers. These are not hypothetical future applications. They are functioning systems, already deployed, already reducing headcount.

JPMorgan’s tools — with names like IndexGPT and Proxy IQ — are handling trading signals and shareholder votes. The bank’s CEO Jamie Dimon has been more direct than most of his peers. “It will eliminate jobs,” he has said flatly. “People should stop sticking their heads in the sand.” He has also called on government, society, and business to discuss retraining and early retirement, acknowledging that the transition will not be painless: “You can’t just take all these people and throw them on the street where the next job is making $30,000 a year, when they’re making $150,000.”

Entry-Level Work Is the First to Go

The heaviest casualties of this first wave are concentrated at the bottom of the professional ladder — exactly where young people are supposed to begin. Back-office compliance work, document processing, basic financial analysis, customer service routing — these are the positions that once served as the training grounds for careers in finance. They are also the positions AI handles most efficiently right now.

JPMorgan Chase received 200,000 applications for roughly 2,000 entry-level positions this year. The mathematical reality embedded in that ratio is sobering. For every hundred people who applied, ninety-nine were turned away — and the industry employing those two thousand is simultaneously developing technology designed to reduce that number further.

Generation Z entered the workforce having been told that education and ambition would be rewarded with opportunity. Jerome Powell and multiple economists have validated that Gen Z is facing a genuine “hiring nightmare,” especially for recent college graduates trying to land their first white-collar job — tied to a low-hire, low-fire labor market, the rapid automation of entry-level roles, and a tech industry whose workforce is getting older as Gen Z’s presence shrinks. It is a generational reckoning that no one in a position of institutional authority seems particularly eager to address with honesty.

The Speed of the Thing

What makes this disruption qualitatively different from prior technological waves is not the fact of displacement — economies have always reorganized around new tools — but the speed. ChatGPT is barely three years old. The major AI platforms that now handle sophisticated financial analysis, legal document review, and software development are younger than most of the entry-level employees they are replacing. The Industrial Revolution unfolded across generations. This one is unfolding across product release cycles.

Goldman Sachs economists have pegged AI-related displacement at between 5,000 and 10,000 monthly U.S. job losses in exposed sectors. Companies are deliberately shifting budgets toward AI infrastructure at the expense of existing headcount — and the spending is not replacing labor gradually. It is replacing it in concentrated bursts tied to specific product decisions and reorganizations. The word “gradual” no longer applies.

The tech sector, often imagined as the beneficiary of the AI revolution, is itself not immune. Block, the payments company founded by Jack Dorsey, laid off more than 4,000 employees — roughly half its total workforce — in 2026. Amazon cut approximately 14,000 positions. Oracle reportedly eliminated between 20,000 and 30,000 roles across multiple regions. These are not the tremors of a sector adjusting. These are institutions fundamentally reorganizing what human labor is for.

The UBI Escape Hatch

When the executives and economists who manage this transformation are pressed about where displaced workers will go, a certain answer has begun surfacing with uncomfortable regularity. Universal basic income — the government providing a direct cash payment to all citizens regardless of employment — is increasingly floated as the long-term equilibrium for a society in which AI handles the majority of productive work. This is the implicit acknowledgment behind the confident earnings calls and the reassuring investor presentations: that there may simply not be enough work left for the number of people who need it.

The political class has not caught up. No serious legislative framework for managing this transition exists at the federal level. The workforce training programs that do exist were designed for industrial dislocation, not algorithmic displacement. And the timeline between “AI is not a threat to your jobs” and “we shed 1,000 positions last quarter” turned out to be less than four months at Bank of America alone.

But UBI is a Marxist abomination, even if some say it is inevitable.

Scripture confronted this tension between the promises of the powerful and the reality experienced by the ordinary long before there were quarterly earnings reports to parse it. In the book of Isaiah, the Lord declares: “Woe unto them that decree unrighteous decrees, and that write grievousness which they have prescribed; to turn aside the needy from judgment, and to take away the right from the poor of my people.” The workers being assured that their jobs are safe while the mechanisms of their replacement are already running deserve something more than corporate candor delivered in hindsight.

What Comes Next

The honest answer is that nobody knows — not the CEOs, not the economists, not the regulators. AI may hit a ceiling. The productivity gains may prove less transformative at higher levels of the career hierarchy than they are at entry level. Consumer benefits — cheaper services, better competition among institutions, more accessible financial products — may eventually create enough economic growth to absorb the displaced.

Or it may not. The models that currently write memos and process creditworthiness assessments are, as of this moment, also being used to write code for better AI models. The feedback loop is not hypothetical. It is running. The labor market is now rewarding specificity over breadth — deep domain expertise paired with AI proficiency, rather than general professional competence alone. That is a meaningful shift in what it takes to remain economically viable, and the window for acquiring those skills is narrowing faster than most public policy discussions have acknowledged.

The executives currently booking record profits while reducing headcount are not villains in a simple morality tale. They are responding rationally to the tools available to them and to the fiduciary obligations they carry. The more uncomfortable question is what obligations exist toward the workers whose livelihoods are being reorganized away — and whether the institutions capable of answering that question have any intention of doing so before the numbers force their hand.

If Bank of America’s Q1 2026 earnings call is any guide, the answer will arrive on a four-month delay, in careful corporate language, after the checks have already cleared.