AI isn't replacing workers. It's redirecting the money that used to pay them

The conventional story about AI and jobs goes like this: A machine learns to do what a human used to do, and the human is let go. That narrative treats displacement as a task-level event. A role is automated, so the role disappears.

But a different mechanism is now visible in corporate earnings calls, SEC filings, and restructuring announcements. Companies are pouring hundreds of billions of dollars into AI infrastructure, and the sheer scale of that investment is creating pressure to find savings elsewhere. Payroll is one of the largest controllable costs, so layoffs function as a way to self-fund the AI buildout without damaging near-term financial results. The worker in this scenario may not have been replaced by an algorithm. But the budget that paid their salary was redirected toward one.

Daniel Keum, a professor at Columbia Business School, described this dynamic in a recent interview with Quartz: Some workers are losing jobs not because their specific roles have been automated but because companies are reallocating resources toward AI and away from everything else.

The distinction matters. It changes how we understand the scale and speed of white-collar displacement.

What companies are saying in their own words

The clearest evidence comes from corporate language itself. "Companies are shifting budgets toward AI investments at the expense of jobs," Andy Challenger, chief revenue officer at outplacement firm Challenger, Gray & Christmas, said in the firm's March 2026 report. "The actual replacing of roles can be seen in technology companies, where AI can replace coding functions. Other industries are testing the limits of this new technology, and while it can't replace jobs completely, it is costing jobs."

That month, AI led all cited reasons for U.S. job cuts, with 15,341 layoffs attributed to the technology, accounting for 25% of total planned cuts. Through the first quarter of 2026, AI ranked fifth among all reasons cited for layoffs at 27,645 cuts, about 13% of the total, but March's jump to 25% suggested an acceleration.

The pattern extends well beyond explicit AI layoffs.

Dell $DELL reduced its workforce by about 11,000 employees in fiscal 2026, bringing headcount to 97,000. The decline represented a 10% year-over-year reduction and marked the third consecutive year the company trimmed its workforce by a similar proportion. In its 10-K filing, Dell described the cuts as part of "disciplined cost management in coordination with our ongoing business modernization initiatives." The workforce adjustments coincided with Dell's push into AI and infrastructure, as revenue in its Infrastructure Solutions Group increased 40% in fiscal 2026, and the company expects AI-optimized server revenue to double in fiscal 2027.

Cisco $CSCO CEO Chuck Robbins was even more direct in 2024 when the company laid off 7% of its workforce. The tech giant was "shifting hundreds of millions of dollars" into AI and other growth areas of its business, Robbins said. Then-CFO Scott Herren framed the cuts not as cost savings but as reallocation, saying to "think of it as reallocating versus being in pursuit of cost savings."

The budget math behind the headcount math

The scale of AI capital expenditure helps explain the pressure on headcount. The five largest U.S. cloud and AI infrastructure providers, including Microsoft $MSFT, Alphabet $GOOGL, Amazon $AMZN, Meta $META, and Oracle $ORCL, have collectively committed to spending between $660 billion and $690 billion on capital expenditures in 2026, according to an analysis by Futurum Group. Among major tech companies, capital expenditures have more than doubled in the last two years, reaching $427 billion in 2025, according to RBC Wealth Management.

That money is going into data centers, GPUs, and networking equipment. It is not translating into hiring. The bulk of spending is going into servers, data centers, and networks, and while those assets are capital-intensive, they are not labor-intensive in the way large manufacturing or services growth might be.

Meta illustrates the tradeoff at its starkest. Meta's 2025 revenues reached $201 billion, meaning the company was not cutting its workforce due to financial strain. The move was about reallocating priorities. Meta set its 2026 capital expenditure guidance at $115 billion to $135 billion, almost double the $72 billion it spent in 2025. To absorb that outlay without collapsing margins, CEO Mark Zuckerberg has been redirecting money that was previously paying salaries. Since 2022, Meta has eliminated about 25,000 positions across multiple rounds of layoffs.

Bank of America $BAC CFO Alastair Borthwick described a quieter version of the same dynamic on the bank's Q4 2025 earnings call. He said the company's headcount was flat during the year, but the expectation was that headcount would continue to come down in 2026. "We can just make decisions not to hire and let the headcount drift down," he said.

Not automation, but appropriation

The distinction between automation-driven displacement and budget-driven displacement has practical consequences. If a worker's role has been automated, the policy response involves retraining for tasks AI cannot do. If a worker's role was eliminated to fund AI infrastructure, the policy response is different: The job itself may still need doing, but the company has decided the capital is better spent elsewhere.

Block $SQ CEO Jack Dorsey cut 4,000 employees in February 2026, and explicitly cited AI in reducing the company's workforce by about 40%. In a post on social media, Dorsey said the cuts were not happening because the business was struggling. He predicted that most companies would reach the same conclusion within a year, saying he would "rather get there honestly and on our own terms than be forced into it reactively."

A March 2026 survey from ResumeBuilder.com of 866 U.S. business leaders found the pattern extends beyond layoffs into compensation: 54% of companies have or will reduce employee compensation to free up capital for AI spending in 2026, and 88% of leaders making these cuts said the weak job market makes it easier to reduce compensation without losing talent. A quarter of S&P 500 companies mentioned at least one quantifiable impact from AI in the first three months of 2026, up from 13% in the same period in 2025, according to a report from Morgan Stanley $MS reported by Axios.

Bank of America said using AI "saves us about 2,000 people" who would typically write code. CEO Brian Moynihan spoke about letting "headcount drift down" thanks to AI.

The pattern is consistent: Companies are not simply automating tasks and then finding they need fewer workers. They are making capital allocation decisions that explicitly prioritize AI infrastructure over headcount, then describing those decisions to investors as strategic discipline.

The workers on the other end of those decisions may find the distinction academic. But for anyone trying to understand the speed of white-collar displacement, the mechanism matters. The threat to many jobs is not that a machine can do them. It is that the money has been sent somewhere else.