AI and its impact on knowledge-intensive entry-level roles

Abstract

This thesis explores how AI-driven automation affects entry-level roles in knowledge-intensive organisations, how organisations adapt onboarding and early-career learning in response, and how expectations about future AI capabilities shape hiring and development decisions. The study draws on 18 qualitative interviews with two groups of Sweden-based actors: knowledge-intensive private sector firms, including managers, HR professionals, recruiters, and consultants; and institutional actors, including public agencies, trade unions, higher education institutions, and business interest organisations. The findings show that AI integration targets codifiable task components rather than entire workflows or occupations, leaving the formal structure of work intact while shifting where entry-level contribution begins. This shift has two compounding consequences: it raises the competence baseline candidates must meet before entering the workforce and simultaneously reduces the learning occasions through which that competence has traditionally been built. Onboarding structures remain formally unchanged, yet the foundational tasks that once developed professional judgement are precisely those now targeted by AI. A recessionary context amplifies both effects, making them difficult to separate empirically. This study of Sweden’s knowledge intensive sector emphasises that the augmentation logic holds, but only for those who have already built the foundational expertise it assumes. At the entry level, that foundation is precisely what AI integration is making harder to build, and neither firms nor institutions responsible for the education-to-work transition have found a reliable way to address what is being lost.