Navigating Governance, Ethics, and Data Security Risks in Artificial Intelligence Adoption
Sathekge M. S. 1 , Bvuma S. 1
1 University of Johannesburg, South Africa
The fast and viral uptake of Generative AI (GenAI) and large foundation models (LFMs) in the corporate worlds is a major, but ill-managed, change in organizational security, risk, and governance. Although GenAI involves an overwhelming number of advantages, its implementation creates a new layer of data security threats that traditional ICT security models were not created to cover. The chapter gives a critical review of the situation in governance today and the particular ethical and technical issues that come along with the integration of GenAI. It is analyzed to explain GenAI Security Threats, such as model poisoning and prompt injection, and the highly significant problem of data leakage and exposure of intellectual property (IP). It also explores the Ethical Gaps, which say that inexplicable bias may yield the results of discrimination or generate shadow vulnerability, which could not be audited. The main contribution is the suggestion of a Socio-Technical Governance Framework that incorporates human control, Explainable AI (XAI), and constant security surveillance into the GenAI deployment pipeline. Actionable Best Practices of data sanitization, model validation and defining clear lines of accountability in AI-driven decisions support this framework. This chapter is meant to inform technology leaders and policymakers by expressing the need to have a proactive risk-based approach to ensure GenAI is exploited safely and in a manner that is responsible in the digital society.
generative AI, data security, corporate governance, ethical AI, risk management, model poisoning, data leakage explainable AI.
APA
Sathekge, M. S., & Bvuma, S. (2026). Navigating governance, ethics, and data security risks in artificial intelligence adoption. In Y. B. Melnyk & M. A. Segooa (Eds.), Artificial Intelligence in Digital Society, Vol. 1. (pp. 118–131). KRPOCH. https://doi.org/10.26697/aids.2026.8
Harvard
Sathekge, M. S., & Bvuma, S. "Navigating governance, ethics, and data security risks in artificial intelligence adoption. In Y. B. Melnyk & M. A. Segooa (Eds.)" Artificial Intelligence in Digital Society, Vol. 1. [online] pp. 118–131. viewed 10 March 2026, https://culturehealth.org/hogokz_knigi/Arhiv_DOI/aids/aids.2026.8.pdf Vancouver
Sathekge M. S., & Bvuma S. Navigating governance, ethics, and data security risks in artificial intelligence adoption. In Y. B. Melnyk & M. A. Segooa (Eds.). Artificial Intelligence in Digital Society, Vol. 1. [Internet]. [cited 10 March 2026]; 118–131. Available from: https://doi.org/10.26697/aids.2026.8 https://culturehealth.org/hogokz_knigi/Arhiv_DOI/aids/aids.2026.8.pdf
Sathekge Machiniba Sylvia – https://orcid.org/0009-0001-9410-3267; Doctor of Business Administration, Doctor, Professor of Practice, University of Johannesburg, Johannesburg, South Africa.
Bvuma Stella – https://orcid.org/0000-0001-8351-5269; PhD in Information Technology Management; Professor, Director, University of Johannesburg, Johannesburg, South Africa.