Schlagwort: research

Veröffentlicht am

[publication] Sieben Mythen der KI-Nutzung #tugraz

Unser Beitrag zu „Sieben Mythen der KI-Nutzung“ hat viele Reaktionen hervorgerufen und nun wurde er auch in die Zeitschrift „Die Österreichische Volkshochschule“ aufgenommen.

Abstract:
Wer das Internet nutzt, kommt im Frühjahr 2026 nicht um Anwendungen generativer Künstlicher Intelligenz (kurz KI) herum. Suchmaschinen bieten neben Links standardmäßig KI-generierte Antworten an, Chatbots unterstützen bei der Buchung von Websites, Schüler:innen lassen sich Tests passend zu den Arbeitsblättern der Lehrer:innen generieren usw. – Doch nicht alles, was uns die KI-Anwendungen liefern, wie wir sie nutzen und ihre Ergebnisse verstehen, ist zutreffend und unproblematisch. Das liegt auch an Missverständnissen darüber, wie KI-Anwendungen funktionieren. Aus unserer Sicht – es gibt dazu noch keine empirische Evidenz – verdienen folgende sieben Aussagen besondere Aufmerksamkeit, insbesondere auch im Kontext von Bildung, Schule und Hochschule:

  1.  KI-Anwendungen sind neutral, objektiv und vorurteilsfrei
  2.  KI-Anwendungen arbeiten logisch
  3. KI-Anwendungen denken und lernen wie Menschen
  4. KI-Anwendungen sind empathisch
  5. KI-Anwendungen sind ökologisch und sozial problemlos
  6. KI-Nutzung ist rechtlich einwandfrei
  7. KI-Anwendungen machen Wissen und Kompetenzentwicklung überflüssig

In diesem Beitrag möchten wir diese als „Mythen“ bezeichneten Aussagen beschreiben und aufzeigen, dass und warum sie nicht zutreffend sind. Damit möchten wir einen zukünftig fundierten Umgang und durch die Beschreibung von KI-Mythen Forschung dazu initiieren und unterstützen. 

Referenz: Schön, S., Brünner, B., Ebner, M., Diesenreither, S., Hanfstingl, B., Krammer, G (2026) Sieben Mythen der KI-Nutzung. Die Österreichische Volkshochschule. Jg. 2026 / 286. [Link]

This is an impactful contributions, methodological rigor, and exceptional novelty in the research field of AI in education.

Veröffentlicht am

[publication] Generative AI Chatbots in Secondary Mathematics Education: Development and Implementation of a Dynamic Large Language Model-Based Learning Assistant for Quadrilaterals #tugraz

Our contribution titled „Generative AI Chatbots in Secondary Mathematics Education: Development and Implementation of a Dynamic Large Language Model-Based Learning Assistant for Quadrilaterals“ is now published.

Abstract:
As artificial intelligence becomes more and more a part of education, the challenge is not about having access to generative tools, but about connecting them with the goals of the curriculum and the needs of the classroom. This chapter presents the design and evaluation of a large language model–based chatbot developed specifically for teaching quadrilaterals in lower secondary mathematics. The chatbot integrates fine-tuning with retrieval-augmented generation (RAG), combining accurate, curriculum-aligned content with flexible, conversational support. The chatbot allows learners to ask conceptual questions, solve problems step by step, receive guided hints, and generate flashcards or exercises of varying difficulty. A hybrid routing mechanism selects the most appropriate response strategy based on user intent. Evaluations using both isolated prompts and multi-turn dialogues demonstrate that the hybrid system significantly outperforms standard LLM baselines in terms of accuracy, consistency, and pedagogical suitability. A classroom trial with 20 students confirmed the tool’s usability and effectiveness; students reported high satisfaction and meaningful engagement. This study demonstrates that, with careful content and architectural structuring, generative AI can enhance student learning while supporting differentiated instruction. Future directions include scaling the approach to other topics and incorporating multimodal capabilities.

[full article @ publisher’s homepage]
[draft @ ResearchGate]

Reference: Mallweger, M., Brünner, B., Ebner, M. (2026). Generative AI Chatbots in Secondary Mathematics Education: Development and Implementation of a Dynamic Large Language Model-Based Learning Assistant for Quadrilaterals. In: Auer, M.E., Nikou, S.A. (eds) GenAI in Novel Educational Applications. Studies in Computational Intelligence, vol 1260. Springer, Cham. https://doi.org/10.1007/978-3-032-16153-6_7

This is an impactful contributions, methodological rigor, and exceptional novelty in the research field of AI in education.

Veröffentlicht am

[presentation] Guiding Innovation with Rapid EdTech User Experience Research Nodes #edmedia #SIG

We also did, as usual, the SIG (Special Interest Group) for Emerging Technologies at the EDMedia Conference. This year the session was titled „Guiding Innovation with Rapid EdTech User Experience Research Nodes„

This presentation is from the Special Interest Group Emerging Technologies for Learning and Teaching from EdMedia 2026 conference workshop on Guiding Innovation with Rapid EdTech User Experience Research Nodes (RETURN). It includes supporting resources for the RETURN Manifesto, selected EdTech case examples, and documentation related to rapid user experience research in educational technology contexts.

[Link to the slides]

This is an impactful contributions, methodological rigor, and exceptional novelty in the research field of AI in education.

Veröffentlicht am

[presentation] Code Meets Career: Employer Interpretations of Coding Bootcamp Graduates in Austria and Germany #research

Another talk we did at this year’s EDMEDIA conference is about the research of Daniela, titled „Code Meets Career: Employer Interpretations of Coding Bootcamp Graduates in Austria and Germany„

Coding bootcamps have expanded rapidly in response to persistent skill shortages in the IT sector and as alternative pathways into the field, particularly for groups underrepresented in technology, such as refugees and women. However, little is known about how these pathways are perceived and valued within organizations, especially in Austria and Germany.
This study therefore examines how companies in Austria and Germany perceive, evaluate, and integrate coding bootcamp graduates. Drawing on a qualitative content analysis of five semi-structured interviews with decision-makers in software development, IT consulting, and human resources, the study focuses on socially situated employer perspectives on skills, potential, and legitimacy, rather than measuring effectiveness in terms of employment outcomes.
The findings reveal an ambivalent picture: while graduates are valued for their motivation, adaptability, teamwork, and practical orientation, concerns persist regarding their theoretical depth and level of project experience. In the case of refugee participants, language proficiency is additionally identified as a key challenge. Overall, the study provides exploratory insights into how coding bootcamp qualifications are interpreted and evaluated by employers, rather than aiming to produce generalizable conclusions.

[Link to the slides]

Veröffentlicht am

[presentation] The OER Paradox in Ukraine #edmedia #research

Another presentation at this year’s EDMedia Conférence is about „The OER Paradox in Ukraine„

This study analyses the impact of the new Law of Ukraine „On Copyright and Related Rights“ (2811-IX) on the use of educational materials in the context of digital transformation and military crisis. Using the benchmarking methodology „15 cases in 15 countries“ and qualitative interviews with teachers, the work compares Ukrainian norms with the practice of EU countries. The results show that broad educational exceptions (in particular, Articles 22 and 24) create a situation of „legal comfort“ for the academic community, allowing the legal use of protected content in closed digital environments. However, this gives rise to the „OER paradox“: the absence of legal barriers to the use of proprietary resources reduces the motivation to create full-fledged open educational resources under free licenses. The paper highlights the need for institutional incentives to overcome dependence on closed content and integrate Ukraine into the global open education movement.

[Link to the slides]

Veröffentlicht am

[publication] Developing and Testing a Peer-Review Process for Content Quality Assurance in MOOCs: A Case Study on an E-Assessment Course #eassessment #tugraz

Our publication about „Developing and Testing a Peer-Review Process for Content Quality Assurance in MOOCs: A Case Study on an E-Assessment Course“ got published now.

Abstract:
This contribution presents the development and testing of a peer-review process for content quality assurance in MOOCs, implemented in the course “E-Assessment – auf Kurs gebracht”. The process was evaluated regarding complexity, duration, collaboration with external reviewers, and learners’ perception. Results show that the procedure can be smoothly integrated into MOOC development. Reviewers contributed beyond expectations by providing materials, didactic advice, and legal-ethical reflections. Learners rated the videos (very) positively (92.7% positive ratings, 100 participants, n = 812 answers), especially for structure and coherence. Slightly lower ratings for ‘visual appearance’ and ‘use of supportive linguistic elements’ can be explained by the course’s retro video design and the viewers’ understanding of how linguistic devices can be effectively used in educational videos. The study confirms peer review as a feasible and effective quality assurance approach that supports both collaboration and content improvement.

[article @ publisher’s homepage]
[draft @ researchgate]

Reference: Loitzenbauer, J., Ebner, M., Schön, S., Brünner, B. (2026). Developing and Testing a Peer-Review Process for Content Quality Assurance in MOOCs: A Case Study on an E-Assessment Course. In: Auer, M.E., Toth, P. (eds) Innovation via Collaborative Learning in Engineering Education. ICL 2025. Lecture Notes in Networks and Systems, vol 1847. Springer, Cham. https://doi.org/10.1007/978-3-032-18885-4_26

Veröffentlicht am

[publiation] AI-Powered Chatbots for Education Using RAGs #tugraz

Our publication about „AI-Powered Chatbots for Education Using RAGs“ got published right now.

Abstract:
This study examines the use of Retrieval-Augmented Generation (RAG) in AI-powered educational chatbots, focusing on embedding efficiency, large language model (LLM) performance, and context integrity. Implemented within the MOOC Informatik-Fit at TU Graz, the system supports scalable, curriculum-aligned self-paced learning. Three embedding models were evaluated, with text-embedding-ada-002 offering the best balance between semantic quality and cost-efficiency. Subsequently, three LLMs—GPT-3.5 Turbo, GPT-4o, and GPT-4o mini—were compared, revealing GPT-4o mini as the most cost-effective option while maintaining high accuracy and contextual coherence. Ethical robustness was assessed using 30 adversarial prompts, demonstrating strong resistance to jailbreaking in both GPT-4o and GPT-4o mini, supporting their suitability for secure and pedagogically reliable MOOC applications.
The paper also presents a replicable framework for the implementation of RAG-based systems, with the objective of promoting personalized, ethical, and accessible digital education on a large scale.

[article @ publisher’s homepage]
[draft @ ResearchGate]

Reference: Brünner, B., Deutschmann, F., Etzelstorfer, S., Lechner, A., Schön, S. & Ebner, M. (2029) “AI-Powered Chatbots for Education using RAGs: A Study on Embedding Efficiency, LLM Performance, and Context Integrity”. In: Transforming Education with Singularity Technologies: Lifelong Learning from Childhood to Adulthood (1st ed.). Uğur, S. (Ed.) Chapman and Hall/CRC. Chapter 9. 21 pages https://doi.org/10.1201/9781003584339

Veröffentlicht am

Blogpost on „New Report: Seven Myths of AI Use – A Critical Perspective on Generative AI“

Thanks to Stefanie, who published a blogpost about our „Seven Myths of AI Use – A Critical Perspective on Generative AI„:

These days, people can hardly use the Internet without running into generative AI—yet many everyday beliefs about “how AI works” are inaccurate in ways that matter especially for education, argues the position paper “Seven Myths of AI Use”. It was published by a team of six Austrian researchers: Sandra Schön, Benedikt Brünner, and Martin Ebner from Graz University of Technology (TU Graz), Sarah Diesenreither and Georg Krammer from Johannes Kepler University Linz, and Barbara Hanfstingl from the University of Klagenfurt. At a time when we start to contemplate futures of “a country of geniuses in a datacenter” (Dario Amodei), this report summarizes some of the most prominent concerns towards generative AI.

Find here the full blogpost.

Veröffentlicht am

[guidlines] A digital textbook for everyone: Accessible and interactive, from concept to use #accessibility #tugraz #research

Our final guidlines „A digital textbook for everyone: Accessible and interactive, from concept to use“ are now online available:

The Guidelines offer practical orientation on how to design digital schoolbooks that are didactically sound, inclusive, and accessible. They address everyone involved in creating or using learning materials — from publishers, authors, and teachers to designers, developers, parents, and educational stakeholders.
A key feature of the project is its transdisciplinary approach: experts from subject didactics, inclusive education, UX design, computer science, and accessibility worked together to develop methods, processes, and examples. While the work is based on content from mathematics and geography, the principles are transferable to other subjects.
Digital schoolbooks provide new opportunities for individualized, collaborative, and competence-oriented learning. The Guidelines show how digital materials can follow a “Born Accessible” strategy from the very beginning, aligned with the principles of Universal Design for Learning. They provide concrete recommendations, best practices, and examples that can be directly applied to development projects.
The content covers the entire development process — from defining inclusive learning goals to decisions about design and representation, technical implementation, feedback structures, and ongoing quality assurance.
These guidelines were developed as one of three key outcomes of the Erasmus+ project Digital Education Material (DEM).

[English version @ Zenodo]
[German version @ Zenodo]

Reference:
English version: Brnic, M., Degenhardt, S., Edelsbrunner, S., Erdel, T., Greefrath, G., Kargl, B., Lathan, H., Macchia, V., Mönter, L., Schön, S., Schütt, M.-L., Torri, S., Wetzel, M., & Ebner, M. (2025). A digital textbook for everyone: Accessible and interactive, from concept to use. CDV. https://doi.org/10.5281/zenodo.18313228
German version: Brnic, M., Degenhardt, S., Edelsbrunner, S., Erdel, T., Greefrath, G., Kargl, B., Lathan, H., Macchia, V., Mönter, L., Schön, S., Schütt, M.-L., Torri, S., Wetzel, M., & Ebner, M. (2025). Ein digitales Schulbuch für alle: Barrierefrei und interaktiv von der Idee bis zur Nutzung. CDV. https://doi.org/10.5281/zenodo.18301620

Veröffentlicht am

[publication] Open Educational Resources for Legal Clarity, Sustainability, and Digital Sovereignty in European University Alliances #tugraz

We summarized the importance of Open Education Resources, especially for Digital Sovereignty. Our paper titled „Open Educational Resources for Legal Clarity, Sustainability, and Digital Sovereignty in European University Alliances“ is now available online.

Abstract:
Open educational resources (OER) are widely recognized for improving access to education and enabling the sharing of knowledge. However, in the context of European university alliances such as Unite!, OER offer additional, often underappreciated benefits that are crucial for cross-border collaboration and sustainable development in higher education. This paper explores three key aspects of OER that are particularly relevant to European alliances. First, OER enable the legally secure use of educational resources across national borders, addressing uncertainties about copyright laws, particularly for translations and adaptations. This ensures compliance with different legal frameworks while fostering collaboration. Second, OER support sustainability by ensuring that investments in educational materials are not limited by restrictive usage rights. This is especially critical in alliances where shared resources are central to fostering long-term cooperation and aligning with sustainability goals, a priority for Unite!. Finally, OER contribute to digital sovereignty by empowering institutions and educators to create, adapt, and share resources without relying on proprietary platforms or licenses. This coincides with European alliances’ broader strategic objective of promoting autonomy and resilience in their digital ecosystems. By highlighting these often-overlooked benefits of OER, the present research aims to broaden the perspective on their strategic importance in fostering collaboration, sustainability, and sovereignty within European university alliances.

[full paper @ publisher’s homepage]
[full paper @ ResearchGate]

Reference: Schön, S., & Ebner, M. (2025). Beyond Open Access: Open Educational Resources for Legal Clarity, Sustainability, and Digital Sovereignty in European University Alliances. Weizenbaum Journal of the Digital Society, 5(4). https://doi.org/10.34669/wi.wjds/5.4.2

This is an impactful contributions, methodological rigor, and exceptional novelty in the research field of Open Educational Resources (OER) and Digital Sovereignty.