AI sparks layoffs and paradoxical warnings of a labor crunch

Companies from Silicon Valley to Wall Street have at times pointed to artificial intelligence as the reason for cutting staff, even as labor economists and industry analysts say broad employment impacts are not yet visible and warn of a different risk: a looming shortage of trained workers for entry-level and midlevel roles.

Vanguard’s analysis, highlighted in recent reporting, examined “roughly 140 occupations it deemed the most vulnerable to getting replaced by AI, including office clerks, typists, HR assistants, law clerks and data scientists.” But Adam Schickling, senior economist at Vanguard, told CNN: “At a high level, we have not seen evidence that AI-exposed roles are experiencing lower employment.” Vanguard concluded that “While AI may have started to change our workflows, its role in explaining the recent slowdown in job growth is overstated.”

That tension—firms saying they are automating tasks while macro data show limited displacement—frames a debate about how and where labor will be affected. Some companies “have recently reported they’re eliminating some positions because AI can automate entry-level workers’ tasks or make current workers more efficient,” CNN reported, yet other analysts caution that the immediate picture is mixed and that longer-term, more targeted shifts are likely.

Industry forecasts point to a concentrated pattern of change. LinkedIn’s outlook for 2026 warns that “Examples of what we are likely to see in 2026 include consolidation of junior analyst teams as AI handles first-pass analysis, reductions in marketing operations and content coordination roles as generative tools absorb execution work, thinner layers of middle management as reporting and status aggregation become automated, and smaller support and operations teams as AI handles intake, routing, and documentation.” LinkedIn says these moves will often be framed as “efficiency initiatives or organizational streamlining rather than explicitly labeled as AI-driven. The underlying mechanism, however, will be task automation.”

That mechanism matters because vulnerability, LinkedIn argues, is defined by task concentration rather than job title. “Roles where day-to-day work is dominated by drafting, summarization, basic analysis, scheduling, or standardized reporting are most exposed,” the company wrote. In many cases, a role may survive while fewer people are required to produce the same output.

The uneven pattern of displacement feeds two competing risks. Some AI executives and researchers have issued stark warnings: “In May, Anthropic CEO Dario Amodei warned AI could eliminate half of all entry-level jobs in white-collar professions, spiking the unemployment rate up to 20% in the near future.” At the same time, corporate leaders such as Cisco President Jeetu Patel caution against retreating from hiring: Patel said it is the “stupidest thing a company can do” in the long run to refuse to hire entry-level workers because of AI and added, “I reject the nation that humans are going to be obsolete in like five years, that we’re not going to have anything to do and we’re going to be sitting on the beach.”

Analysts and policy commentators are pushing a third path: prepare the labor force. Democracy Journal posited that “an interesting possibility is that while college graduates could face more replacement by AI, workers without college degrees might now become more productive,” citing examples where AI could help mechanics diagnose machinery or nursing assistants assess patients. It warned, however, that “some workers are, in fact, hurt by automation” and that “displaced workers can suffer large and lasting losses, sometimes not regaining employment.” Its policy prescriptions are explicit: “Governments at all levels can improve education and training options to make workers more complementary with AI,” promote AI literacy and human skills in K-12 and higher education, and expand experiential and apprenticeship programs to replace lost entry-level channels.

For policymakers and corporate leaders, the immediate task is pragmatic: monitor task-level displacement, preserve entry-level pipelines, and scale training that pairs humans with AI. For workers and voters, the central question is whether governments and employers will act quickly enough to prevent the loss of careers even as firms pursue efficiency gains.