Artificial intelligence is starting to hit the job market where companies traditionally hire first, not through broad-based layoffs but by shrinking the need for junior white-collar staff and shifting more work toward oversight of software tools. In recent remarks to Axios, Anthropic Chief Executive Dario Amodei said AI could eliminate “half of all entry-level white-collar jobs” within one to five years, a warning that sharpened a debate already building across boardrooms, campuses and labor economists’ models.
The pressure is showing up in early-career data before it appears in economywide payrolls. The Federal Reserve Bank of New York said in its latest labor-market analysis that recent graduates in computer engineering and computer science faced unemployment rates above several other majors, with computer engineering near 7.5% and computer science around 6.1%. In a statement accompanying its release, the New York Fed said labor conditions for new graduates “deteriorated noticeably” in several technical fields, a notable reversal for disciplines long treated as the safest path into the professional class.
That mismatch is becoming harder for elite schools to ignore. Fortune reported that a course developer at a top business school said students already arrive with extensive hands-on exposure to AI models, adding, “Our faculty are passionate, but there are two problems,” including the speed at which the technology outpaces curriculum design. The comment, cited by Fortune, captures a broader concern among educators and employers: universities still market AI-era opportunity, while companies increasingly say they need fewer entry-level workers doing routine analytical tasks.
Corporate leaders are framing the shift less as a layoff story than as a redesign of work. At the World Economic Forum in Davos earlier this year, ServiceNow Chief Executive Bill McDermott said, “If we have hired ‘nines and tens,’ why should we fire them instead of re-tooling them?” according to public remarks from the event. McDermott said employees whose prior work centered on repetitive IT tasks increasingly move into supervising and orchestrating AI agents, a model many software and services companies now present as the preferred route to productivity gains without the reputational and operational costs of mass cuts.
Others have been more explicit about the labor tradeoff. Salesforce Chief Executive Marc Benioff said in recent public comments that AI now handles a large share of customer-service interactions, reducing the need for some support roles while increasing demand for workers who can manage complex escalations and higher-value client relationships. In interviews reported by outlets including Bloomberg and CNBC, Benioff has argued that digital labor is changing staffing models across sales, service and software development, even as the company continues to hire in selected AI-focused areas.
The consulting industry sees a substantial medium-term impact, though not an immediate collapse in employment. In a recent outlook, Boston Consulting Group estimated that 10% to 15% of current jobs could be displaced by 2030 or shortly thereafter as companies deploy more autonomous AI systems. BCG said the effect is likely to fall hardest on routine and rules-based work, while roles requiring judgment, client management and cross-functional decision-making could expand. That distinction matters for executives because it suggests organizational charts may flatten at the bottom even if total payrolls do not plunge.
Bank economists and academic researchers are also urging caution against sweeping claims. Goldman Sachs economists have said generative AI has the potential to automate a meaningful share of tasks across administrative and professional occupations, while also lifting productivity and creating new categories of work over time. Separately, a recent working paper circulated through the National Bureau of Economic Research found that most firms surveyed reported little measurable effect from AI on employment or productivity so far. Co-author John Haltiwanger said the evidence suggests adoption remains uneven, with many companies still experimenting rather than fully redesigning operations around the technology.
That unevenness helps explain why labor-market disruption remains concentrated in hiring pipelines instead of headline-grabbing job cuts. Employers can often absorb AI efficiency by slowing recruitment, leaving open positions unfilled and asking existing staff to work with new tools. Economists cited by Reuters and The Wall Street Journal in recent reporting have said this pattern tends to hit recent graduates first because entry-level jobs often involve the very documentation, coding assistance, research support and customer-response tasks that AI systems now perform at low cost.
The implications extend beyond technology companies. If junior roles become scarcer across finance, consulting, marketing, legal services and back-office operations, businesses may save money in the near term but risk weakening the training ground that produces future managers and specialists. That concern is already surfacing in executive discussions about reskilling and internal mobility. What comes next will matter more than the current headlines: investors will watch whether AI spending translates into sustained margin gains, policymakers will track whether graduate unemployment spreads beyond tech, and companies will need to prove they can automate entry-level work without hollowing out the talent pipeline they still depend on.
JBizNews Desk
