The Practice of Reflective Teaching with AI
Future Skills Development with Offline LLMs
DOI:
https://doi.org/10.83397/PAIHE.2026.02Keywords:
generative AI, reflective teaching, future skills, offline large language models, AI literacy, constructive alignmentAbstract
This practice report describes how generative AI was integrated by reflective teaching into a course on values-oriented corporate communication at the end of a bachelor’s program in media management in the summer term of 2026. It documents a first teaching cycle under a minimal consensus for designing university teaching with generative AI, which is set out in the foundational contribution to PAIHE; the course followed an earlier version of that consensus. Students worked in groups on responsibility analyses and values-oriented communication strategies. The group process was designed so that students first noted their own assessments and then used several large language models, including models running offline on their own devices, to add new perspectives and roles; students documented and evaluated their AI use in their presentations, and this documentation entered the assessment. The report explains the didactic rationale, the selection of tools that non-specialists can install and run locally, and the AI rules of the course. From the instructor’s perspective, the work products showed unusual depth and quality. These are observations of work products, not evidence of competence; whether the design builds disciplinary and AI-related judgment competence is a question for further teaching cycles.
Practical AI competence and understanding was enhanced by continuous use and reflection of usefulness. The report closes with lessons for the next cycle, including the need to shorten and adapt content.
References
Baker, R. S., & Hawn, A. (2022). Algorithmic bias in education. International Journal of Artificial Intelligence in Education, 32(4), 1052–1092. https://doi.org/10.1007/s40593-021-00285-9
Bastian, M. (2024, February 2). Nvidias „Chat with RTX“ ist ein kostenloser, lokaler Daten-Chatbot [Nvidia’s “Chat with RTX” is a free, local data chatbot]. heise online. https://www.heise.de/news/Nvidias-Chat-with-RTX-ist-ein-kostenloser-lokaler-Daten-Chatbot-9627934.html
Bauduin, S. (2023, August 16). KI statt Gaming: Nvidia soll 2023 mehr als 500.000 HPC-Beschleuniger absetzen [AI instead of gaming: Nvidia expected to sell more than 500,000 HPC accelerators in 2023]. PC Games Hardware. https://www.pcgameshardware.de/Kuenstliche-Intelligenz-Hardware-279517/News/Nvidia-verkauft-eine-halbe-Million-KI-Beschleuniger-in-2023-1426781/
Biggs, J. (1996). Enhancing teaching through constructive alignment. Higher Education, 32(3), 347–364. https://doi.org/10.1007/BF00138871
Biggs, J., Tang, C., & Kennedy, G. (2022). Teaching for quality learning at university (5th ed.). Open University Press.
Digital Education Council. (2025, March 3). Digital Education Council AI literacy framework. https://www.digitaleducationcouncil.com/post/digital-education-council-ai-literacy-framework
Hailikari, T., Virtanen, V., Vesalainen, M., & Postareff, L. (2022). Student perspectives on how different elements of constructive alignment support active learning. Active Learning in Higher Education, 23(3), 217–231. https://doi.org/10.1177/1469787421989160
Herrmann, D. (2026). KI-Policy Presets [AI policy presets]. https://ki-policy.org
Hiltscher, J. (2023, November 28). 500.000 KI-GPUs in einem Quartal sind noch zu wenig [500,000 AI GPUs in one quarter are still not enough]. Golem. https://www.golem.de/news/h100-a100-500-000-ki-gpus-in-einem-quartal-sind-noch-zu-wenig-2311-179820.html
Long, D., & Magerko, B. (2020). What is AI literacy? Competencies and design considerations. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (pp. 1–16). ACM. https://doi.org/10.1145/3313831.3376727
Lubbe, A., Marais, E., & Kruger, D. (2025). Cultivating independent thinkers: The triad of artificial intelligence, Bloom’s taxonomy and critical thinking in assessment pedagogy. Education and Information Technologies, 30(12), 17589–17622. https://doi.org/10.1007/s10639-025-13476-x
Moorhouse, B. L., Yeo, M. A., & Wan, Y. (2023). Generative AI tools and assessment: Guidelines of the world’s top-ranking universities. Computers and Education Open, 5, Article 100151. https://doi.org/10.1016/j.caeo.2023.100151
Mühling, T., Warnke, S., & Bolsinger, H. J. (2026). Reflektierte KI-Nutzung in Lehrveranstaltungen: Befunde aus dem Pretest am Ende des Sommersemesters 2026 [Reflective use of AI in university courses: Findings from a pretest at the end of the 2026 summer semester]. Practice of Artificial Intelligence in Higher Education, 1(1), Article 3. https://doi.org/10.83397/PAIHE.2026.03
Technische Hochschule Würzburg-Schweinfurt. (2024, July 1). Technische Hochschule Würzburg-Schweinfurt führt hochschulweit ChatGPT ein [Technical University of Applied Sciences Würzburg-Schweinfurt introduces ChatGPT university-wide]. https://www.thws.de/service/news-presse/pressemeldungen/thema/technische-hochschule-wuerzburg-schweinfurt-fuehrt-hochschulweit-chatgpt-ein/
Warnke, S., Bolsinger, H. J., & Mühling, T. (2026). Leistung ist nicht Lernen: Ein Minimalkonsens für die Gestaltung von Hochschullehre mit generativer KI [Performance is not learning: A minimal consensus for designing university teaching with generative AI]. Practice of Artificial Intelligence in Higher Education, 1(1), Article 1. https://doi.org/10.83397/PAIHE.2026.01
Weidlich, J., Gotsch, F., Schudel, K., Marusic-Würscher, C., Mazzarella, J., Bolten, H., Bütler, D., Luger, S., Wohlfender, B., & Maag Merki, K. (2025). Teacher, peer, or AI? Comparing effects of feedback sources in higher education. Computers and Education Open, 9, Article 100300. https://doi.org/10.1016/j.caeo.2025.100300
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Harald Bolsinger, Sven Warnke, Tanja Mühling (Autor/in)

This work is licensed under a Creative Commons Attribution 4.0 International License.