The rapid proliferation of generative Artificial Intelligence (GenAI) in our life and work presents a dual challenge for Vocational Education and Training (VET): educators should utilise AI for pedagogical efficiency and improvement while simultaneously preparing students for labor markets where AI is increasingly embedded in professional practice. Despite this "dual significance" of AI as both a teaching and learning tool and a vocational competency (Attwell et al., 2020), there is a paucity of research on how VET teachers acquire and should acquire the necessary literacy to integrate these tools effectively.

This paper presents the results of a pilot study conducted within an Erasmus+ cooperation partnership involving four European teacher educator universities. The VETAssIst project (2024-2027, No. 2024-1-HU01-KA220-VET-000253387) developed and piloted a specialised e-learning course designed to enhance VET teachers' AI literacy. The partners designed the curriculum and learning content in eight modules focusing on different aspects of the VET teaching work (e.g., GenAI for active learning, GenAI for assessment, GenAI for planning, GenAI for professional development etc.), developing VET students’ AI literacy, and ethics and data protection. Each module consists of a Lesson, a practical Assignment, and questions encouraging self-reflection (AI Learning Diary) and the exchange of experiences in asynchronous group discussions (Forum). By aligning the modules with the competencies outlined in an AI Supplement to DigCompEdu (Bekiaridis &  Attwell, 2024), UNESCO’s AI framework (Miao, F., & Cukurova, 2024), and the cognitive levels of Bloom’s Taxonomy (Oregon State University, 2024), we aimed to create practical and pedagogically sound learning experiences. The course was implemented in Moodle in five language versions with slightly adapted and supplemented national language content to suit the local needs. The pilots were organised in slightly different ways in the four partner countries but using a uniform core set of questions to collect participants’ feedback to finalise the curriculum and learning content. This study thus adopts a Design-Based Research (DBR) approach, characterising the educational intervention as an iterative process enacted in real-world settings.

The theoretical framework distinguishes between AI as a mere technical tool for routine automation and AI as a tool for pedagogical augmentation (Huang & Rust, 2021). While early adoption often focuses on efficiency (Celik et al., 2022), our study also aimed to explore the deeper transformative potential of AI in VET. We define this potential specifically as how and in what aspects AI can improve VET teaching - moving beyond mere workload reduction to enhance active and collaborative learning, personalisation, and the creation of adaptive learning environments that were previously difficult to implement (Kasneci et al., 2023).

The primary objective of this study is to evaluate the pilot implementation of the VETAssIst course. We seek to answer the following research questions:

1. How effectively does the VETAssIst course support AI literacy development among VET teachers?

2. What specific barriers (technical, linguistic, or contextual) do VET teachers encounter when integrating GenAI tools into classroom and workshop settings and other aspects of the teaching work?

3. In what aspects and ways can AI improve or transform VET teaching?

The study posits that successful AI adoption in VET requires a curriculum that is responsive to the specific "timetable realities" of VET educators and addresses the ethical and social risks unique to the sector (UNESCO, 2023).

The study employs a Convergent Mixed Methods Design, collecting quantitative and qualitative data simultaneously throughout the pilot phase to provide a holistic evaluation of the VETAssIst course. The pilot involved 53 VET teachers (including 6 student teachers) from four partner countries (Hungary, Germany, Serbia, Slovakia), selected to represent a diverse range of VET subjects and digital skill levels. Data collection was embedded directly into the learning process to capture immediate reflections and behavioral patterns: Quantitative Instrument: - Feedback Forms: Short, closed-question surveys administered pre-course, post-module, and post-course via Moodle. These measured satisfaction with content and technical usability. Qualitative Instruments: - Focus Group Discussions: Online or face-to-face meetings were conducted in each partner country. A core set of semi-structured questions was utilised to explore the strengths and weaknesses of the VETAssIst curriculum and learning content, and the potential and deep-seated challenges of using AI for improving VET teaching. - AI Learning Diaries: Participants were requested to answer questions prompting self-reflection on learning processes and "lightbulb moments" regarding AI application throughout the course. - Forum Analysis: Asynchronous discussions on the Moodle platform, prompted by tutor-designed questions, were analysed to identify shared challenges and peer-to-peer learning dynamics. The analysis follows a triangulation protocol. Quantitative data from Moodle forms provides descriptive statistics on course satisfaction and usability. Qualitative data from transcripts of focus groups, diaries and Forum posts are subjected to thematic analysis.

The study anticipates two primary categories of results: Curricular Refinements and Pedagogical Insights. Regarding curricular refinement, the analysis will yield a validated version of the VETAssIt course. We expect to resolve the tension between providing a broad "buffet" of GenAI tools versus deep-diving into a select few. Feedback on the assignments will determine if practical, artifact-creation tasks drive higher engagement than theoretical reflection. Pedagogically, the study will report on the transformative potential of AI in VET as perceived by practitioners. We anticipate findings that detail how GenAI may improve VET teaching by enhancing active learning, enabling differentiation in heterogeneous student groups and simulating complex professional scenarios in the classroom. Furthermore, the results will highlight the shift in teacher mindset, from viewing AI as a "shortcut" to understanding it as a partner for pedagogical innovation. Ultimately, the findings will inform VET decision-makers at both school and national levels, outlining how institutional and policy support must evolve to sustain AI literacy. The final output will be an open-access, empirically validated course tailored to the unique needs of the VET sector.

Attwell, G., Deitmer, L, Tutlys, T., Roppertz, S., & Perini, M. (2020). Digitalisation, artificial intelligence and vocational occupations and skills: What are the needs for training teachers and trainers? In C. Nägele, B. E. Stalder, & N. Kersh (Eds.) Trends in vocational education and training research, Vol. III. Proceedings of the European Conference on Educational Research (ECER), Vocational Education and Training Network (VETNET), 30–42. https://doi.org/10.5281/zenodo.4005713 Bekiaridis, G., & Attwell, G. 2024. Integrating Artificial Intelligence in Vocational and Adult Education: A Supplement to the DigCompEdu Framework. Ubiquity Proceedings, 4(1): 20. https://doi.org/10.5334/uproc.142 Bükki, E., Papp, Z., Manojlovic, H., Sanitizer, A., Luprihová, J., & Kovács, E. (2025). Artificial Intelligence in VET: Interest and concerns by VET teachers. In: VETNET Conference Series, 8 (ECER), 59-67. https://doi.org/10.21240/vetcon/2025/ecer/32 Celik, I., Dindar, M., Muukkonen, H., & Järvelä, S. (2022). The Promises and Challenges of Artificial Intelligence for Teachers: a Systematic Review of Research. TechTrends 66, 616–630. https://doi.org/10.1007/s11528-022-00715-y Kasneci, E., et al. (2023). ChatGPT for Good? On Opportunities and Challenges of Large Language Models for Education. Learning and Individual Differences, 103, 102274. Manojlovic, H., & Bükki, E. (forthcoming). Artificial Intelligence in VET: Interests and concerns by VET teachers. In: International Conference on Interactive Collaborative Learning 2025, Springer. Miao, F., & Cukurova, M. (2024). AI competency framework for teachers. UNESCO Digital Library. https://unesdoc.unesco.org/ark:/48223/pf0000391104 Pelletier, K., McCormack, M., Reeves, J., Robert, J., Arbino, N., Dickson-Deane, C., & Stine, J. (2022). 2022 EDUCAUSE Horizon Report Teaching and Learning Edition. Boulder, CO.: EDUCAUSE https://library.educause.edu/-/media/files/library/2022/4/2022hrteachinglearning.pdf?la=en&hash=6F6B51DFF485A06DF6BDA8F88A0894EF9938D50B Oregon State University: Bloom’s taxonomy revisited. https://ecampus.oregonstate.edu/faculty/artificial-intelligence-tools/blooms-taxonomy-revisited/ (2024), accessed 26 January 2026. UNESCO (2023). Guidance for generative AI in education and research. https://doi.org/10.54675/EWZM9535