Responsible generative AI enablement for learning futures in higher education: a qualitative integrative review and bounded institutional illustration

Abstract

Purpose This paper aims to examine how higher education institutions can move from fragmented generative artificial intelligence (GenAI) experimentation toward responsible institutional enablement for learning futures. Design/methodology/approach A qualitative integrative review was reconstructed through a retrospective audit of a 48-record working library, reconciliation with sources already used in the manuscript and a targeted August 2026 update. Forty-one sources were retained and synthesized using a multilevel lens combining technology-adoption theory with a relational socio-technical perspective. A bounded institutional illustration was mapped to the resulting provisional framework. Findings The synthesis identifies six recurring institutional capability domains: governance, faculty capability-building, pedagogical redesign, research enablement, infrastructure and access and evaluation. These domains are organized into a provisional Responsible Institutional Enablement Framework. The institutional illustration documents capacity-building activities and descriptive reach, but not behavioral, educational or institutional outcomes. Practical implications Institutions can use the provisional framework to audit gaps, sequence capability-building and design evaluation plans while adapting expectations to local resources, governance and disciplinary contexts. Originality/value The contribution lies in integrating literature streams that are commonly separated, distinguishing capacity-building activity from measured outcomes and making the evidentiary status of the proposed framework explicit. The framework is offered as a review-derived organizing synthesis requiring independent validation, not as a validated or wholly novel theory.