Year: 2026
Type: review
Study design: Systematic Review
Abstract: This study conducts a systematic literature review, following the PRISMA 2020 protocol across Scopus and Web of Science, yielding 36 articles.It examined how GenAI integration in frontline service operations redefines task automation boundaries, reconfigures work and competency demands, and gives rise to new sociotechnical governance arrangements.The findings reveal that GenAI dissolves the classical routine/non-routine boundary established by Autor et al. (2003), entering cognitive and socio-emotional service interaction domains, with automation operating along a complexity-determined gradient formalized through the Switch Point concept.A structural upskilling-deskilling paradox emerges simultaneously: while GenAI improves service outcomes and generates new competency demandssuch as prompt engineering or AI trainersit can erode domain expertise through overreliance across healthcare, financial services, education, and hospitality contexts.Addressing this tension requires governance at three interdependent levels: operational (HITL configurations and RAG-anchored systems), organizational (AI Teams and new professional roles), and institutional (Corporate Digital Responsibility frameworks).Collectively, these findings contribute a gradient conception of automation boundaries, consolidate the upskilling-deskilling tension as a structural cross-sectoral phenomenon, and propose a three-level governance architecture for frontline service delivery.Competitive advantage stems not from GenAI adoption per se, but from the organizational capability to configure and govern human-AI collaboration.
Language: Portuguese
Subfield: Safety Research
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7225065199