Work

A framework for the secure adoption of generative artificial intelligence in the banking sector

Year: 2026
Type: article
Abstract: The integration of Generative Artificial Intelligence (GenAI) into the banking sector presents significant opportunities for innovation, particularly in enhancing operational efficiency, customer engagement, and risk management. However, the secure adoption of GenAI remains a critical concern, especially within DevOps environments that underpin continuous software development and deployment. This study develops a framework for the secure adoption of GenAI in DevOps within the South African banking sector. Guided by the Nelson Mandela University Design Science Framework Methodology (NMU-DSFM), the research follows a four-phase process: problem analysis, solution development, refinement, and validation. A comprehensive literature review and stakeholder engagement inform the identification of core security, governance, and regulatory requirements. The proposed framework is iteratively refined and validated through structured interviews and scenario-based simulations using AI personas. Key evaluation criteria include regulatory alignment, governance and risk management, threat modelling, secure DevOps integration, and scalability. The findings confirm that the framework effectively addresses the unique challenges of GenAI deployment in banking DevOps pipelines, offering a practical and adaptable solution for secure implementation. This research contributes to the growing body of knowledge on AI governance and provides actionable guidance for financial institutions seeking to responsibly integrate GenAI technologies.
Language: English
Open Access status: green
OpenAlex ID: W7224693250
FWCI:
Cites: 75
Cited by:
Related to: 10