Artificial Intelligence for Cybersecurity: Governance Insights from Cybersecurity Experts in the Banking Industry in South Africa

Abstract

Artificial Intelligence (AI) has emerged as a transformative force in cybersecurity, particularly in the banking sector, where cyber threats continue to evolve rapidly. South African banks, operating in a highly regulated environment, face the dual challenge of leveraging AI to enhance cybersecurity while ensuring compliance with governance and ethical standards. This study investigated the governance mechanisms required for the responsible adoption of AI for cybersecurity within the South African banking sector. Guided by the Technology Acceptance Model (TAM), Innovation Diffusion Theory (IDT), and Institutional Theory, this qualitative research explored cybersecurity experts' perspectives on AI adoption, governance, and ethical concerns. A single-case study design was employed, involving in-depth interviews with ten certified professionals from a major South African bank. Thematic analysis was conducted using Atlas.ti to identify key themes in AI governance. Findings revealed that while AI supports enhanced threat detection, response efficiency, and fraud prevention, notable challenges persist—including integration with legacy systems, skills shortages, regulatory ambiguity, and ethical risks. Experts highlighted the need for transparency, ethical oversight, and regulatory alignment to mitigate the potential harms of AI-driven cybersecurity decisions. Organisational readiness, leadership support, and structured governance frameworks were found to be critical enablers of responsible AI implementation. This study contributes to the evolving AI governance discourse by proposing an AI for Cybersecurity Governance Framework tailored to the South African banking sector. The framework offers strategic insights for banks, regulators, and policymakers seeking to balance innovation with ethical and regulatory imperatives in cybersecurity.