Challenges, safeguards, and professional learning needs for AI integration: insights from a two-week AI literacy module with preservice teachers

Preservice teachers (PSTs) often have limited preparation for integrating artificial intelligence (AI) in pedagogically meaningful and ethically responsible ways, while teacher education programs face practical constraints in providing sustained, practice-based AI preparation. This exploratory single-case study examined how 29 PSTs participating in a two-week, ethics-embedded AI literacy module within a required technology integration course anticipated and/or experienced challenges, proposed and/or enacted safeguards, and identified professional learning supports for responsible AI integration. Using a qualitatively driven mixed-methods design, data were drawn from reflections, structured AI tool evaluations, group lesson plans, and a post-module questionnaire. Findings showed that PSTs’ challenges clustered around ethical, pedagogical, and technical issues. Across artifacts, PSTs most readily translated concerns into visible, teacher-led safeguards related to data protection, verification of AI outputs, bounded use, teacher mediation, and attribution or critique, but were less consistent in operationalizing safeguards related to bias, fairness, and AI’s instructional authority. PSTs also identified ongoing needs for professional learning related to ethical use, technical fluency and prompting, foundational AI understanding, and curriculum-aligned integration. The findings offer design-relevant insight into how AI-related concerns surfaced in PSTs’ evaluative and instructional design work, including which safeguards were more readily operationalized and which required more support.