Framing Data Governance Amid AI Advancements in a Public University in South Africa

Abstract

Data governance (DG) has emerged as a critical domain within information technology, especially in light of growing data volumes, regulatory pressures, and institutional performance demands. In higher education institutions (HEIs), the shift toward data-driven decision-making is intensifying amid increasing accountability, financial constraints, and the rapid digital transformation of administrative and academic functions. DG encompasses the policies, processes, roles, and technologies that inter-alia ensures data quality, consistency, compliance, oversight and security across the institutional landscape. However, many universities continue to face challenges with fragmented data systems, inconsistent data definitions, data silos, and underutilised data assets—issues that inhibit institutional effectiveness and strategic planning. This study proposes a structured Data Governance Framework (DGF) tailored to the unique context of South African universities. Drawing on a mixed-methods approach anchored in design science research methodology, data was collected through a DG maturity assessment survey and qualitative focus group sessions with key university stakeholders. The findings informed the design of a framework for the governance of data which is sensitive to the specific context while addressing governance gaps, roles and responsibilities, quality management, continuous improvement, change management, standardisation and regulatory alignment.Crucially, this study situates data governance within the emergent challenges and opportunities presented by Artificial Intelligence (AI) and Large Language Models (LLMs). As HEIs begin to adopt AI-powered tools for administration, research, and teaching, the need for robust DG becomes more urgent. The proposed framework incorporates AI-readiness and ethical oversight mechanisms to ensure trustworthy data handling, algorithmic transparency, data privacy, and responsible innovation in an AI-augmented environment.The resulting DGF provides a structured approach for universities to manage and safeguard their data assets, reduce institutional risk, foster data trust, and optimise data value—while aligning governance practices with the ethical imperatives introduced by AI and LLM technologies.