The Vanishing Apprenticeship: Expertise Debt and the Disappearing Verifier in AI-Augmented Organisations

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Organisations are delegating entry-level work to artificial intelligence while asking experienced staff to verify what these systems produce. This paper argues that the two movements can be connected in a way that creates a hidden liability. The judgment needed to verify work is renewed, in large part, through correction episodes: a person attempts meaningful work, receives specific feedback from someone able to tell the difference, and reasons differently afterwards. Where AI absorbs the work that hosted those episodes and nothing replaces them, an organisation may continue to consume verification capability while ceasing to renew it. We propose expertise debt as a construct for the resulting future shortfall and describe a condition, the disappearing verifier, in which strong current reviewers conceal a weak successor pipeline. Whether the eventual loss is gradual, abrupt or absent is treated as an empirical question; we hypothesise sharper deterioration where review is concentrated in a few people with weak backup. The paper contributes a mechanism, four testable propositions, the Apprenticeship Diagnostic, and the Verifier Continuity Matrix, a proposed framework that separates current verification capability from successor pipeline strength. It links verifier continuity to AI decision authority through an impact and reversibility lens and sets out an assessment approach for executive education. The evidence reviewed also shows that AI can support learning, so the risk depends on work design rather than on the technology alone. The argument is conceptual, the frameworks are not empirically validated, and the propositions are offered for testing.