For years the worry was robots on the factory floor. The data in 2026 tells a different story: the industries absorbing the deepest and fastest AI-driven disruption right now are ones that sit behind desks.
Goldman Sachs Research said in an April 24, 2026 note that AI-related substitution has wiped out roughly 25,000 jobs per month over the past year, while AI augmentation has added about 9,000, for a net drag of about 16,000 a month. Goldman has also argued that the negative effects are falling largely on younger, less-experienced workers. That is a net monthly reduction of about 16,000, and it is not slowing.
Where the Damage Is Deepest
The widely circulated “FAIR Framework” displacement scores cited here are not from a generally recognized research institution, and the underlying methodology is not presented in a way that can be independently verified. The directional takeaway still holds: the roles most exposed to near-term AI substitution tend to be task-heavy desk jobs, especially administrative support, routine finance operations, customer support, and standardized marketing work.
Financial services have seen reductions in some back-office functions as automation expands across transaction processing, fraud detection, compliance monitoring, and parts of customer support. Goldman Sachs Research has also highlighted that jobs such as telephone operators, insurance claims representatives, and bill collectors are among those at higher risk from AI substitution. For traders, that matters beyond the human-interest angle: firms cutting back-office headcount are compressing their cost base in ways that flow directly to operating margins.
Goldman’s own labor-market work has also pointed to early signs in the data: it said there is evidence that employment in information and communication services, two of the industries most exposed to AI, has slowed because of AI, and that employment in these sectors has slowed across nearly all major developed markets since 2022. Tech-layoff totals for 2026 vary widely depending on the tracker and definitions used, and public datasets do not support a single, definitive “160,000” figure with “82% in the US” and “nearly half tied directly to AI restructuring.”
The Entry-Level Trap
The structural problem hiding inside these aggregate figures deserves attention. Stanford HAI’s 2026 AI Index highlighted that employment among software developers aged 22 to 25 has fallen nearly 20% since 2024, even as older colleagues’ headcount grows. Entry-level white-collar roles have traditionally been the first rung on the career ladder, and the research suggests that rung is getting wobbly. If companies use AI to handle junior tasks, the pipeline that trains future senior staff could thin out.
Customer service is also shifting fast. McKinsey has noted that self-service channels can already handle 70% to 80% of interactions in many service models, and that more advanced AI and digital setups can push that share much higher. Separately, McKinsey Global Institute has estimated that today’s technology could, in theory, automate about 57% of current US work hours, and Brookings has estimated that about 6.1 million workers, primarily in clerical and administrative roles, face both high AI exposure and low capacity to adapt.
The compliance layer adds another dimension to this shift. European regulators are now stress-testing whether financial institutions can exit cloud providers without triggering systemic risk, a pressure that compounds the AI-driven restructuring already reshaping bank back offices.
The Sectors That Are Holding
Not every corner of the market is in retreat. In healthcare, AI can assist with clinical work, but it does not eliminate the need for clinicians and in-the-room judgment. Where healthcare looks most exposed in the near term is the administrative layer, including billing, coding, scheduling, and documentation, and that is exactly where AI is cutting first.
An IMF staff discussion note published in January 2026 found that roughly one in ten job vacancies in advanced economies lists at least one new skill, often appearing first in the United States. The IMF also noted that AI-related skills are associated with higher wages, while warning that the gains can be uneven and may deepen labor-market polarization.
The trading implication is straightforward. Firms in financial services, insurance, and enterprise software that are actively cutting headcount through AI tools are not in distress. They are compressing labor costs on a schedule the market has not fully priced into forward multiples. The sectors absorbing the pain on the workforce side are the same sectors where operating leverage is about to become a more interesting story than revenue growth. That is where the attention belongs this quarter.
The operating-leverage story has a structural accelerant behind it. Enterprise software consolidation is compressing SaaS spend as mega-cap platforms absorb point solutions, amplifying the margin gains that AI-driven headcount cuts are already delivering.
