IMF Warns AI Will Eliminate Entry-Level Jobs, Trapping Gen Z

Entry-level white-collar jobs are vanishing faster than new roles can replace them, creating a structural trap for young workers entering the global labor market. International Monetary Fund Managing Director Kristalina Georgieva characterized the unfolding disruption as a “slow-moving tsunami,” warning that the traditional career ladder is breaking at its lowest rungs.

The IMF estimates that roughly 40% of jobs worldwide face augmentation, downgrading, or outright elimination as artificial intelligence capabilities expand. The immediate casualties are not manual laborers but recent graduates who historically relied on routine, repetitive office tasks to build foundational experience. Without those stepping stones, Georgieva noted during the IMF and World Bank annual meetings in Bangkok on Monday, young professionals have no clear path to advance up corporate ranks.

This dynamic threatens to hollow out middle-skill employment, particularly in advanced economies where administrative, analytical, and clerical roles are most exposed to automation. Workers possessing specialized AI skills are commanding higher wages, yet that premium has not translated into broad-based hiring. Instead, labor demand is bifurcating: high-paying technical roles on one end, and lower-wage physical jobs such as restaurant work and data center construction on the other.

“Burst the myth that AI is shrinking demand for people,” Georgieva said. “What it is doing is shifting demand for people now. That may not be the case in the future.”

Singapore offers an early indicator of how quickly this shift can materialize in highly developed, tech-forward economies. President Tharman Shanmugaratnam revealed that only 74.4% of university graduates secured full-time positions in 2025, a sharp drop from 79.4% the previous year, according to data reported by the Straits Times.

Shanmugaratnam urged governments to treat the declining placement rate as a long-term structural failure rather than a temporary cyclical dip. He pointed to the aggressive workforce interventions deployed after the 2008 financial crisis and the COVID-19 pandemic as blueprints for stabilizing labor markets before they fracture. When graduates fail to secure initial employment, they do not simply pause their development. They compete directly with the next graduating class, compounding the surplus while losing intangible professional attributes.

“We can’t afford to lose that because if you lose that, you not only have that cohort of young people waiting a year and then competing with the next cohort the next year, you lose more than one year of skills,” Shanmugaratnam said. “You lose a whole set of intangible attributes, and you lose trust in the future.”

That erosion of confidence is already triggering political instability. In France, hundreds of thousands of students recently took to the streets to protest deteriorating educational infrastructure and demand increased funding, reflecting deepening anxiety among younger demographics about their economic prospects.

Governments facing these demographic pressures must also prepare for severe fiscal consequences. Widespread youth underemployment reduces consumer spending and depresses tax revenues precisely when states need capital to fund retraining programs and modernize education systems. Georgieva argued that nations can no longer afford educational models built on the assumption that workers will learn a single skill set and apply it for decades.

Not all technology leaders share the IMF’s urgency regarding near-term displacement. OpenAI CEO Sam Altman conceded in May that the pace of entry-level job destruction has lagged behind his own projections.

“I’m delighted to ⁠be wrong about this,” Altman said during an interview with Commonwealth Bank of Australia CEO Matt Comyn. “I thought there would have been more impact on entry-level white-collar jobs being eliminated by now than has actually happened.”

Anthropic CEO Dario Amodei, who previously cautioned that AI could wipe out half of all white-collar positions, has since outlined a theory where automation compresses existing roles rather than deleting them entirely. Under his model, if software handles 90% of a worker’s tasks, the remaining human responsibilities expand to fill the role, theoretically multiplying individual productivity tenfold.

U.S. President Donald Trump has maintained a similarly optimistic posture, recently suggesting the term “super intelligence” for the technology. Speaking to the New York Times in January, he predicted the sector would ultimately generate massive employment growth.

“I think AI is going to be a tremendous job producer,” Trump said, adding that the economy would eventually face a shortage of human workers, necessitating widespread robotic deployment.

Despite these counterarguments, current labor data aligns more closely with the warnings issued in Bangkok. Recent U.S. employment reports have shown persistent softness in sectors like professional and business services, which traditionally absorb college graduates. Temporary help services—often considered a leading indicator for broader hiring trends—have experienced consecutive months of contraction throughout 2026, suggesting employers remain hesitant to commit to expanding junior headcounts even as enterprise software spending reaches record highs.