The rapid emergence of generative artificial intelligence (GenAI) technologies such as ChatGPT, Microsoft Copilot, and Canva AI is driving profound changes in educational leadership and administration. This qualitative study examines how Slovenian school leaders perceive and use GenAI in their work and which opportunities, risks, and support needs they identify. Drawing on four focus groups across primary, secondary, and higher education within the national project ”Generative artificial intelligence”,, we used a qualitative methodology; explicitly a reflexive thematic analysis in ATLAS.ti. We identifed five main themes: (i) pragmatic, low-risk adoption focused on administrative efficiency; (ii) uneven AI literacy and demand for hands-on professional development; (iii) ethical and legal concerns (GDPR/EU AI Act) around privacy, authorship, and transparency; (iv) infrastructural and cost constraints limiting high-stakes use; and (v) emergent opportunities for data-informed decision-making and individualized support. We theorize these patterns using Diffusion of Innovations and Technology Acceptance perspectives to explain why administrative use predominates and pedagogical integration remains tentative. Findings show that GenAI is predominantly used for streamlining administrative work, document preparation, and routine communication. Its application in teaching and learning remains limited and experimental. Participants highlighted both the benefits of GenAI such as increased efficiency, support for creative work, and potential for individualized learning and a range of persistent challenges, including gaps in digital literacy, concerns over data privacy and academic integrity, and unequal access to advanced technologies. The study draws on Diffusion of Innovations, the Technology Acceptance Model, Sensemaking Theory, Educational Change Theory, and Distributed Leadership to explain how school leaders interpret and implement GenAI in practice. We offer actionable implications for leadership development, assessment redesign, and policy aligned with the EU AI Act. The contribution is a leadership-centred, EU-contextualized account that clarifies conditions under which GenAI currently shifts not replaces leadership practice.
Flogie et al. (Sat,) studied this question.