Artificial Intelligence (AI) could undermine the traditional route into professional careers by allowing companies to automate the routine tasks through which young workers acquire experience, India’s Chief Economic Adviser V. Anantha Nageswaran warned at the 2026 Per Jacobsson Lecture.
Nageswaran said companies were hiring fewer junior employees because AI could perform work that previously justified the cost of recruiting and training graduates.
The consequences could extend beyond immediate job losses, leav ing businesses with fewer experienced workers capable of supervising AI systems in the future.
In his remarks at the IMF-World Bank Meetings in Bangkok, Nageswaran cited a study covering 41 countries in which the share of junior workers declined in 23 of the 31 countries with measurable data.
In the United States, employment among workers aged 22 to 25 in occupations most exposed to AI had fallen about 19% behind that of older workers in the same occupations, he said, citing research presented during the lecture.
Nageswaran argued that the economic incentive to automate entry-level tasks could create a collective problem. Individual companies may save money by reducing junior recruitment, but the wider economy could lose the training opportunities needed to develop future specialists.
The issue is particularly important for developing economies, where large numbers of young people need access to employment and opportunities to build professional skills.
Clark acknowledged that AI could disrupt employment markets, saying its effects would vary across occupations. Customer service and call-centre roles could face job losses, while some scientific disciplines might experience higher productivity and potentially increased employment.
He said Anthropic was supporting an initiative called Claude Corps, intended to create 1,000 jobs for recent graduates working with nonprofit organisations in the United States. The company would pay their salaries, he said, presenting the programme as one experiment in helping young people enter the workforce.
Clark also called for better employment data, including information on hiring and training by seniority, so governments could assess how AI was changing labour markets.
Nageswaran said policymakers should pay greater attention to occupations requiring physical presence, practical skills and human judgment, including nursing, teaching, electrical work, plumbing and hospitality.
These roles could become more valuable as AI makes some forms of cognitive work cheaper, he argued.
The debate highlights a policy challenge: AI could increase productivity while weakening the systems through which workers gain experience. Without new approaches to recruitment, apprenticeships and training, short-term efficiency gains could come at the expense of the next generation of skilled workers.
