According to a study published in Frontiers in Education, pre-service teachers who perceive stronger institutional support for artificial intelligence also report more responsible digital citizenship practices, with AI self-efficacy playing a mediating role in this relationship. The research, conducted among 618 pre-service teachers in Oman, is among the first to examine these three factors together in a single model within teacher education, particularly in an Arab Gulf context.

The study, authored by researchers from Sultan Qaboos University, Ain Shams University, the Ministry of Education in Oman, and Tanta University, was grounded in organizational support theory and social cognitive theory. The researchers used a cross-sectional survey and covariance-based structural equation modeling to test their hypothesized mediation model.

The findings showed that perceived institutional support for AI was positively associated with both AI self-efficacy and digital citizenship practices. AI self-efficacy, in turn, was positively associated with digital citizenship practices. The indirect association between institutional support and digital citizenship practices through AI self-efficacy was statistically significant and consistent with partial mediation, the authors reported.

The study defines digital citizenship practices as responsible, ethical, and professional conduct with technology — an increasingly recognized component of professional digital competence for teachers. The authors argue that teacher preparation is a formative period during which enduring technology-related beliefs and practices are established, making it a critical window for shaping how future teachers model technology use for their students.

For university students studying to become teachers, the findings suggest that the institutional environment — including resources, infrastructure, policies, training, technical assistance, and ethical guidance — may matter as much as individual motivation in fostering responsible AI use. The study implies that teacher education programs should not only teach about AI but also create conditions that build students' confidence and sense of institutional backing.

The research was conducted in Oman, where national policy frameworks such as Oman Vision 2040 and the National Program for Artificial Intelligence and Advanced Digital Technologies (approved in September 2024) have positioned AI and digital transformation as central to economic diversification and institutional development. The General Policy for the Safe and Ethical Use of Artificial Intelligence Systems, adopted in 2025, further emphasizes human-centered principles including fairness, transparency, accountability, and privacy.

The authors noted that existing research on AI in higher education has largely focused on individual-level factors such as attitudes, acceptance, literacy, and adoption intentions, while ethical and institutional dimensions remain underexplored. They also pointed to a geographic concentration of studies in Asia, North America, and Western Europe, with limited evidence from Arab Gulf contexts despite ambitious national AI agendas in the region.

As a cross-sectional survey, the study establishes associations but cannot determine causation. The authors acknowledged this limitation and called for longitudinal and experimental designs to further test the causal pathways suggested by their model. The sample was drawn from one country, which may limit generalizability to other settings.

For student teachers and education policymakers, the study reinforces the idea that responsible AI engagement is not solely an individual responsibility but depends on coordinated institutional strategies. The authors concluded that their findings offer implications for teacher education programs seeking to cultivate competent, ethically informed, and responsible AI engagement among future educators.