This thesis examines the legal and strategic framework for artificial intelligence (AI) in the public sector of Slovenia and the European Union, assessing how regulatory and organizational solutions shape AI’s practical deployment. Methodologically, it relies on qualitative analysis of legislation and strategies, complemented by a comparative review of key domains of use—e-government, health and social care, education and research, justice, and smart cities as well—and selected national practices. At the EU level, the Artificial Intelligence Act (Regulation (EU) 2024/1689), built on a risk-based approach, stands alongside the GDPR to define obligations for high-risk systems and to strengthen transparency and trust. In Slovenia, AI deployment is framed by ZVOP-2, ZUP, and ZDIJZ, as well as strategic documents such as the National AI Programme and Digital Slovenia 2030. The findings indicate that AI offers operational benefits (automation of routine tasks, more efficient processing, and more tailored services) while raising issues of ethics, privacy, explainability, and accountability; successful scaling depends on institutional capacity, data standards, and workforce skills. The thesis contributes a comprehensive comparative perspective and practical recommendations for responsible, human-centred AI adoption in public services that couples regulatory safeguards with innovation.
|
