Ethical Frameworks and Governance for Generative Artificial Intelligence in Public Services: A Multidimensional Analysis

The public sector is coming into its own as new opportunities for Generative Artificial Intelligence (GenAI). It can help to streamline tasks, enhance interactions with citizens, and aid administrative processes. Meanwhile, there are significant issues of bias and privacy, transparency and accountability and absence of human judgment in day-to-day decisions. While there are technical solutions, there is no workable governance model in the public sector that meets its needs. This is the purpose of this research: to fill in this gap and offer a multidimensional approach to governance of GenAI in public services. It reviews 12 major AI governance frameworks, which were published between 2018 and 2026, and are based on the public values framework of Bozeman, the concept of digital bureaucracy of Bovens and the administrative values of Hood. The analysis is related to the technical, institutional and ethical aspects of these materials that are combined. Based on the results, it is found that the benefits of GenAI can only be realized with a Three-Dimensional Governance Model, which is a Layered Responsibility Framework between regulators, GenAI developers, public organizations and frontline officials. Then, the model is extended to civil administration, health and education. At the heart of the recognition is a recognition of long-term practical protection (such as the right to intervene in automated systems), the right to appeal decisions made by an AI and requiring human oversight. These safeguards build on each other and provide a credible alternative to ensure that GenAI supports the public interest and is held accountable and trusted.

Generative Artificial Intelligence (GenAI), Algorithmic Accountability, Public Values, Ethical Governance, Public Services

1. Abuzir, Y. (2026). Guiding pedagogical safety in the digital age: An implementation of the Human Artificial Intelligence Rights (HAIR) model in public education. International Journal of Educational Technology in Higher Education, 23(1), 89–104. [Google Scholar] [Crossref]

2. Ansell, C., & Gash, A. (2008). Collaborative governance in theory and practice. Journal of Public Administration Research and Theory, 18(4), 543–571. [Google Scholar] [Crossref]

3. Bovens, M. (2007). Analysing and assessing accountability: A conceptual framework. European Law Journal, 13(4), 447–468. [Google Scholar] [Crossref]

4. Bovens, M., & Zouridis, S. (2002). From street-level to system-level bureaucracy: How information and communication technology copes with discretion. Public Administration Review, 62(2), 174–184. [Google Scholar] [Crossref]

5. Bozeman, B. (2007). Public values and public interest: Counterbalancing economic individualism. Georgetown University Press. [Google Scholar] [Crossref]

6. Carrasco, L. B. (2025). Towards responsible AI: A conceptual framework for mitigating ethical risks in generative artificial intelligence systems. AMCIS 2025. [Google Scholar] [Crossref]

7. Deng, M., Yang, R., Zheng, X., Deng, Y., & Jiang, J. (2025). Artificial intelligence in diabetes care: From predictive analytics to generative AI and implementation challenges. Frontiers in Endocrinology, 16, Article 1620132. https://doi.org/10.3389/fendo.2025.1620132 [Google Scholar] [Crossref]

8. Goyal, R., Deshmukh, S. G., & Bolia, N. (2026). Navigating barriers to GenAI adoption in public administration: A systematic evaluation and policy roadmap. Socio-Economic Planning Sciences, 105, Article 102461. https://doi.org/10.1016/j.seps.2026.102461 [Google Scholar] [Crossref]

9. Hood, C. (1991). A public management for all seasons? Public Administration, 69(1), 3–19. [Google Scholar] [Crossref]

10. Kumar, P., & Kumar, S. (2026). Designing trustworthy artificial intelligence: Ethical frameworks, transparency and accountability in AI systems. Journal of Trust Research, 14(1), 45–63. [Google Scholar] [Crossref]

11. Liu, J. Y. (2025). Generative artificial intelligence: Ethical challenges and trust mechanisms. IT Professional, 27(2), 34–41. [Google Scholar] [Crossref]

12. Mashaw, J. L. (1983). Bureaucratic justice: Managing social security disability claims. Yale University Press. [Google Scholar] [Crossref]

13. Mwogosi, A. (2025). Ethical and privacy challenges of integrating generative AI into EHR systems in Tanzania. Digital Health, 11, 23–38. [Google Scholar] [Crossref]

14. Ofosu Asare, Y. (2026). Guiding the future: Developing an ethical framework for generative AI use in education. International Journal of Information and Learning Technology, 43(1), 56–72. [Google Scholar] [Crossref]

15. Scott, W. R. (2014). Institutions and organizations: Ideas, interests and identities (4th ed.). SAGE Publications. [Google Scholar] [Crossref]

16. Tu, J., & Tang, M. (2025). The ethical reconstruction of generative AI: Social value, governance challenges and pathways to digital empowerment. Lecture Notes in Computer Science, 14500, 210–226. [Google Scholar] [Crossref]

17. Torraco, R. J. (2005). Writing integrative literature reviews: Guidelines and examples. Human Resource Development Review, 4(3), 356–367. [Google Scholar] [Crossref]

18. Whittemore, R., & Knafl, K. (2005). The integrative review: Updated methodology. Journal of Advanced Nursing, 52(5), 546–553. [Google Scholar] [Crossref]

• Smart Iot Device for Weather And Health
• Merlarchive: A Web-Based Academic Hub for UDM With AI-Powered Natural Language Processing
• Enhanced Social Network Security System: Integrating Biometric Authentication for Improved User Verification and Privacy Protection
• Smart Budget Allocation in Public Policy: A Data-Driven Approach for Equitable Resource Distribution
• Decision Support System for Faculty Selection, Promotion, and Reclassification Using Predictive Analytics