A comprehensive new study from the McKinsey Global Institute has found that AI is displacing entry-level and junior knowledge-worker jobs at a rate approximately three times faster than economists projected just two years ago. The research, which surveyed 1,200 companies across 15 industries and 40 countries, paints a sobering picture for recent graduates entering the workforce.
Most Affected Roles
The study identifies five categories of work most at risk in the near term: data entry and processing, basic content creation, initial customer service triage, junior software testing, and routine legal and financial document review. In companies that have deployed AI tools aggressively, headcount in these categories has fallen by an average of 31% over 18 months, while output remained stable or increased.
Particularly striking is the data on entry-level professional roles. Law firms using AI contract review tools have reduced the number of first-year associates assigned to due diligence tasks by an average of 40%. Financial services companies using AI for initial financial analysis report needing fewer analysts at the junior level to produce the same volume of reports.
What’s Not Being Replaced
The study is careful to note that AI is not reducing total employment in most industries — rather, it’s changing the composition of the workforce. Roles involving complex client relationships, original strategic thinking, physical presence, and interdisciplinary judgment are growing. The problem is that these roles typically require experience that workers traditionally gained in entry-level positions that are now disappearing.
“We’re seeing a hollowing out of the career ladder,” said Dr. Kweilin Ellingrud, a co-author of the report. “Workers need entry-level experience to reach mid-level roles, but those entry points are being automated. This creates a structural problem we haven’t fully grappled with yet.”
Geographic Variation
The impact is not uniform globally. Countries with high labor costs and strong digital infrastructure — the United States, United Kingdom, Germany, South Korea, and Japan — show the fastest displacement. Developing economies with lower labor costs are experiencing slower AI adoption in these roles, giving workers more time to adapt, but researchers caution this is likely a delay rather than immunity.
Policy Recommendations
The report calls on governments to accelerate investment in retraining programs, consider adapting education systems to emphasize uniquely human skills, and evaluate tax policies that may inadvertently incentivize replacing workers with AI. Several economists quoted in the study advocate for targeted tax credits for companies that invest in worker retraining alongside AI adoption.