Risks and benefits posed by artificial intelligence to workers’ mental health: a systematic literature review

Purpose The purpose of this study is to systematically review the existing literature to explore how the adoption of artificial intelligence (AI) influences workers’ mental health (WMH). Grounded in the complementarity and substitution view (CSV) of technology, the review seeks to identify the potential benefits and the challenges posed by AI, offering a balanced and integrative understanding of its psychological impact on workers. Design/methodology/approach This paper adopts a systematic literature review (SLR) approach, following PRISMA guidelines. A total of 438 articles were initially retrieved from Scopus and Web of Science databases. After a rigorous screening process based on inclusion and exclusion criteria, 60 articles were deemed eligible for in-depth analysis. Findings AI impacts WMH in beneficial, risky and ambiguous ways. Benefits include decision-making support, job satisfaction, efficiency, performance and accuracy. Risks involve deteriorated workplace relationships, loss of autonomy, privacy concerns and unreliability. Ambiguous effects relate to mental health dynamics, skills and workload. Some of such effects impact WMH directly, while others configure as indirect pathways. Originality/value Despite AI’s growing workplace presence, its effects on WMH are underexplored. This systematic literature review fills that gap by synthetizing current knowledge through complementarity and substitution lenses, outlining that both complementarity and substitution views can lead to risks and advantages for WMH. It reveals a research focus on healthcare, with psychology and business fields less studied, and offers theoretical insights, practical implications and future research directions on the direct and indirect effects of AI on WMH.