Abstract: Generative AI has been increasingly integrated into educational settings for its potential to facilitate personalized learning. However, evidence-based implementation concepts remain essential to realize its pedagogical value. Therefore, this study investigates the educational efficacy of ADA, a domain-specific AI chatbot customized for secondary mathematics education, in enhancing learning outcomes and student engagement. A cluster-randomized controlled study was conducted with 195 ninth-grade students from German secondary schools in authentic classroom settings. The objective of the study was to compare ADA's personalized support against conventional differentiation materials. The intervention targeted the Heron method for estimating square roots in a single-lesson intervention with a pre-post assessment. As one of the few cluster-randomized controlled studies on custom AI chatbots in secondary mathematics education, this investigation assessed both affective and cognitive aspects of learning simultaneously to provide a comprehensive research picture. The findings indicated a high degree of student acceptance across the dimensions of the Technology Acceptance Model. Additionally, situational interest significantly improved in the chatbot condition. The study found that trends in performance and emotional responses were favorable, while cognitive load increased slightly. The findings emphasize the evident potential of customized AI chatbots to supplement differentiated instruction in the domain of mathematics education. They provide a critical foundation for future research investigating the potential of AI chatbots to enhance student learning through careful integration and customization.

Abstract: As Artificial Intelligence in Education (AIEd) transforms teaching and learning with accompanying technological, pedagogical, and ethical challenges, understanding pre-service teachers’ AI perceptions and ensuring their adequate preparation is crucial for effective AIEd implementation. While previous research has examined pre-service teachers’ AI competencies and perspectives using either quantitative or qualitative methods, a critical gap remains in understanding their specific needs for AI-based instructional tools and relevant training through an integrated theoretical lens. This mixed-methods study addresses this gap by investigating 49 pre-service teachers’ needs under the Intelligent-TPACK framework. Statistical and thematic analyses revealed pre-service teachers’ ambivalent attitudes toward AI-based tools and associated training. Participants expressed needs for technological, pedagogical, content, and ethical knowledge related to AI in education, along with concerns about AI-based tools. While individual requirements for AI-relevant knowledge varied, participants consistently demonstrated high demand for training specifically in AI ethics. Overall, this study revealed pre-service teachers’ unpreparedness regarding AIEd, furtherly uncovering critical gaps between their knowledge demands and existing teacher training programs. The findings call for an integrated approach combining AI-technical expertise with hands-on pedagogical practices within teacher education programs. This research contributes to the field by validating the Intelligent TPACK framework and providing recommendations for educational program designers to create effective training for AI-based tools.

Abstract: • Revealed progression of AI pedagogical competencies linked to academic seniority among pre-service teachers in Latin America. • Found negligible gender and age differences in AI literacy, underscoring equitable digital competence distribution. • Provided a psychometrically robust, culturally adapted tool for measuring AI-enhanced TPACK, fostering AI-ready teacher training. Pre-service teachers can play a crucial role in integrating AI-based tools into the new educational landscape. However, there is a need to validate specialized instruments, apply current conceptualizations such as intelligent-TPACK, and address ethical issues, as pre-service teachers are often overlooked in the development of tools for AI integration. To address these gaps, we adapted a previously existing instrument designed for in-service teachers to measure pre-service teachers’ integration of AI within their training context. We conducted a quantitative cross-sectional survey with a total of 366 pre-service teachers to evaluate the adapted intelligent-TPACK instrument and examine participants' demographic characteristics related to the framework dimensions. Data analysis included a Confirmatory Factor Analysis to assess the factor model of the adapted instrument, followed by correlations to compare participant variables such as gender, type of university, and stage in the training program with the Intelligent-TPACK model factors. To investigate the differences among groups, the nonparametric ANCOVA test (Quade test) was utilized, enabling the control of covariates like age and academic progress level to ensure comparability across the dimensions of the Intelligent-TPACK model. Findings reveal a high fit of the Intelligent-TPACK model for pre-service teachers (CFI=.997; TLI=.997). The data also shows statistically significant effects related to academic progress level and type of institution, while factors -gender, geographic location, and type of major- did not demonstrate noteworthy differences. These results highlight key areas for future curriculum development and support for pre-service teachers in integrating AI education.

Abstract: This study explores the perceptions and experiences of secondary school teachers in Albania who implemented Content and Language Integrated Learning (CLIL) combined with digital tools, particularly EdPuzzle, in teaching natural sciences subjects during and after the COVID-19 pandemic (online and face to face). The research focuses on understanding the long-term impact of these methodologies on teaching practices and their potential for continued use in an ever-changing educational landscape. A phenomenological approach was employed to gather in-depth insights through semi-structured interviews, capturing thirty language and natural sciences teachers' reflections on the benefits, challenges, and future possibilities of integrating CLIL with video-based learning tools. Using phenomenological approach, the findings reveal that the teachers observed significant improvements in student engagement, comprehension, and autonomy through the use of CLIL and EdPuzzle. However, they also highlighted challenges related to technological access and the need for ongoing professional development. The study concludes that while the innovative teaching methods fostered during the pandemic have the potential for sustainable use, their success hinges on continued support and adaptation to the evolving educational context in Albania. Recommendations include the integration of CLIL into broader curriculum design, the provision of resources and training for teachers, and further research into the scalability of these practices across different educational settings.

Abstract: As generative artificial intelligence (GenAI) rapidly becomes a structural element of education, teacher preparation programs face urgent challenges in developing both pedagogical competence and ethical awareness among future educators. This exploratory study investigates how Swiss pre-service teachers in business education conceptualize, design, and evaluate GenAI-supported instruction through the lens of the Intelligent-TPACK framework, an expanded model integrating technological, pedagogical, content, and ethical knowledge. Twelve master-level pre-service teachers participated in a mixed-method study that combined self-report surveys, group-based instructional design artefacts, and AI-driven prompt analysis using OpenAI o3. Results show that participants reported high confidence in technological and pedagogical AI knowledge, but they exhibited weaker confidence and reliability in ethical knowledge, particularly regarding transparency and accountability. All chatbot designs address the "Active" rather than "Interactive" level of the ICAP hierarchy. The AI-based analysis further highlighted gaps in Socratic questioning and metacognitive prompting, underscoring limited opportunities for reflection and co-construction. These findings reveal a need to move beyond surface-level tool familiarity towards integrating ethical reflection and explicit design-in-action practices within teacher education. Ultimately, the study underscores the importance of cultivating pre-service teachers as co-designers of pedagogical experiences who are equipped to navigate both the technical and ethical complexities of AI-mediated classrooms.

Abstract: In today ’ s era of technological evolution, programming education is crucial for shaping the future workforce and fostering innovation. However, access to quality computer science education remains a significant challenge with Sub-Saharan Africa nations experiencing a pronounced digital divide. Despite growing interest in technology, these countries struggle with unequal access to educational resources. AI-driven tools like ChatGPT, Codey, and GitHub Copilot offer personalized learning experiences that could democratize access to knowledge and reshape programming education. This quantitative study examines the impact of these AI tools on fostering inclusive education in Kenya, Nigeria, and South Africa. It involves 322 university students, using purposive sampling and online questionnaires. Various quantitative analyzes, including descriptive statistics, country-wise comparisons, one-way ANOVA, Kruskal – Wallis tests, and correlation analysis, were conducted. The study reveals students ’ motivations for programming, their attitudes towards AI-driven educational tools, and the perceived impact on equity, diversity, and inclusion. Significant variations were found in attitudes based on educational level and country of residence, highlighting the need for tailored strategies to enhance the inclusivity and effectiveness of AI-driven programming education tools.