[presentation] Synthetic Educators: Analyzing AI-Driven Avatars in Digital Learning Environments #HCII25

Our research about AI-Driven Avatars was presented at 27th International Conference on Human-Computer Interaction, Gothenburg, Sweden.

Struger, P., Brünner, B., & Ebner, M. (2025, Juni 23). Presentation: Synthetic Educators: Analyzing AI-Driven Avatars in Digital Learning Environments. Graz University of Technology. https://doi.org/10.3217/bngt5-p2053

[publication] Early Findings from Pilots in AI-Driven Education: Effects of AI-Generated Courses and Videos on Learning and Teaching #research #tugraz #AI

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

Our chapter in the new book on Emerging Technologies, „Early Findings from Pilots in AI-Driven Education: Effects of AI-Generated Courses and Videos on Learning and Teaching,“ has been published.

Abstract:
This paper presents the (preliminary) findings from three pilot activities conducted at TU Graz over the past 18 months, exploring the use of AI in generating (open) educational resources. The first pilot involved developing a MOOC titled “Societech: Society in the Context of Information Technologies”, which utilized various AI tools to create videos and learning materials, engaging over 500 participants. The second pilot focused on creating multilingual videos for the MOOC “Open Educational Resources in Higher Education”, using AI-generated avatars of course instructors, involving more than 800 participants. The third pilot was a field study conducted in an Austrian secondary school, where 20 students aged 12–15 could choose between videos featuring their teacher, an AI-generated human avatar, or a cartoon character. Preliminary results indicate a clear preference for teacher-led videos, highlighting the significant role of the teacher in the learning process. While AI tools facilitated quicker and more cost-effective production of educational resources, challenges such as the need for quality assurance and handling of (now possible) automatic translations were noted.

[draft @ ResearchGate]
[full version @ publisher’s website]

Reference: Schön, S., Brünner, B., Ebner, M., Edelsbrunner, S., Hohla-Sejkora, K., Uhl, B. (2025). Early Findings from Pilots in AI-Driven Education: Effects of AI-Generated Courses and Videos on Learning and Teaching. In: Auer, M.E., May, D. (eds) 2024 Yearbook Emerging Technologies in Learning. Learning and Analytics in Intelligent Systems, vol 44. Springer, Cham. https://doi.org/10.1007/978-3-031-80388-8_2

[publication] Rapid Multilingual Video Production utilizing Artificial Intelligence #AI

At this year’s EDMedia conference, we published a publication titled „Rapid Multilingual Video Production utilizing Artificial Intelligence„.

Abstract:
This study presents the Rapid Multilingual Video Production (RMVPro) framework, an optimized process for creating high-quality multilingual educational videos using artificial intelligence (AI). The framework involves three stages: planning, translation, and production, with each stage involving iterative feedback loops with experts and native speakers to ensure linguistic and cultural accuracy of the translations. The framework for the Multilingual Video Sprint was tested during an international conference and the produced multilingual videos showing an AI human avatar were evaluated in a MOOC with feedback from 128 students. Results showed that AI could be utilized to accelerate video production while maintaining the clarity and comprehensibility of educational content. However, feedback indicated mixed perceptions of AI-generated human avatars, with some students preferring human facilitators for reasons of authenticity. The use of AI avatars in multilingual educational videos must be implemented carefully, but the RMVPro framework ensures high quality while utilizing the potential of AI to rapidly produce multilingual educational videos.

[draft @ researchgate]
[full version @ conference website]

Reference: Brünner, B. & Ebner, M. (2025). Rapid Multilingual Video Production utilizing Artificial Intelligence. In T. Bastiaens (Ed.), Proceedings of EdMedia + Innovate Learning (pp. 1223-1230). Barcelona, Spain: Association for the Advancement of Computing in Education (AACE). Retrieved May 28, 2025 from https://www.learntechlib.org/primary/p/226280/.

[video] Digitale Bildung #avatar #ki #tugraz

Wir haben wieder etwas mit „meinem“ Avatar herumgespielt und ein Video erstellt rund um das Thema „Digitale Bildung“. Das gibt es nun in zwei Versionen – deutsch und englisch. Vielleicht ist es ja auch in anderen Kontexten hilfreich, freu mich wie immer auf Rückmeldungen.

YouTube

Mit dem Laden des Videos akzeptieren Sie die Datenschutzerklärung von YouTube.
Mehr erfahren

Video laden

Version in deutscher Sprache
YouTube

Mit dem Laden des Videos akzeptieren Sie die Datenschutzerklärung von YouTube.
Mehr erfahren

Video laden

Version in englischer Sprache

[presentation] Building OER Competence Across the Unite! Alliance. Insights from the „Unite! OER Courses“ Project #OER #tugraz #research

For the 2nd Unite!-Ed Future Conference, we presented the results of our OER course, titled „Building OER Competence Across the Unite! Alliance. Insights from the „Unite! OER Courses“ Project

The „Unite! OER Courses“ seed fund project successfully advanced Open Educational Resources (OER) competence among students and lecturers across the Unite! alliance. A major milestone was the development of a Massive Open Online Course (MOOC) on OER, licensed under CC BY 4.0 International and translated into 11 languages, including Turkish and Indonesian, with the help of AI avatars and tools. Since its launch on May 6, 2024, the MOOC has attracted over 1,200 participants. Additionally, an OER course in English was introduced on the Unite! federated learning management system Metacampus.
To further enhance capacity, a dedicated OER training for lecturers was conducted, which resulted in the publication of a collection of tools and methods for future OER training initiatives. In this contribution, we build upon previous work, particularly insights from our project report (Unite, 2024) and various presentations on the topic, including preliminary findings on the impact at partner universities (Schön et al., n.d., submitted for JODDE).
Methodologically, we addressed different research questions primarily through a case study, allowing us to analyse the implementation, reception, and institutional effects of the OER initiatives within the alliance. The collected data provides valuable insights for the establishment of a future Unite! Open Science Academy.
This presentation will share the project’s achievements, methodologies, and outcomes while addressing challenges and opportunities for promoting OER practices in international higher education settings.

Find the slides here:

Schön, S., Ebner, M., & Vicente-Saéz, R. (2025, März 13). Building OER Competence Across the Unite! Alliance. Insights from the „Unite! OER Courses“ Project. Presentation at Unite! Ed Future Conference, March 13, 2025. Graz University of Technology. https://doi.org/10.3217/hm2aw-dvh27

[interview] Using AI-Generated Instructor Video for Multilingual Content in MOOCs #tugraz #edtech

Benedikt gave an interview about „Using AI-Generated Instructor Video for Multilingual Content in MOOCs“ to the „Association for the Advancement of Computing in Education“ and introduced our recent research on AI-genereated videos:

Benedikt Brünner is a member of the e-education team at TU Graz, and facilitates the production of avatars, assets, audio and video. As part of his PhD project he has conducted initial survey research to gauge students’ reactions to AI avatars. While most students found the avatars to be authentic and natural, they still preferred real human teachers over AI avatars. However, many students indicated that they would not have realized that the videos were AI-generated, which shows the increasing sophistication of the technology.
In the interview, Benedikt Brünner discusses the potential of generative AI for creating educational videos.

[Link to the interview]

[publication] Workshop: Creating Educational Videos with an AI Avatar Video Generator #tugraz #research

We did a workshop at this year’s ED-Media conference in Brussels about „Creating Educational Videos with an AI Avatar Video Generator

Abstract:
The integration of Artificial Intelligence (AI) into the creation of learning materials represents a cutting-edge frontier. This workshop is designed for educators, content creators, and instructional designers, and will introduce participants to the potential of AI avatar software in the production of educational videos. Through interactive demonstrations and guided practice, participants gain practical skills in utilizing AI avatar software to create educational videos that meet different learning styles and needs. We will explore case studies of successful implementations and discuss strategies for integrating AI avatars into various educational contexts, from K-12 to higher education and corporate training. Additionally, the workshop will address ethical considerations and best practices in the use of AI in education, ensuring that participants are equipped to responsibly navigate the challenges of digital content creation. Participants will leave the workshop with an initial understanding of the capabilities of AI avatar software and their first experience in creating educational videos with this new technology.

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

Reference: Brünner, B. & Ebner, M. (2024). Creating Educational Videos with an AI Avatar Video Generator. In T. Bastiaens (Ed.), Proceedings of EdMedia + Innovate Learning (pp. 1165-1166). Brussels, Belgium: Association for the Advancement of Computing in Education (AACE). Retrieved July 22, 2024 from https://www.learntechlib.org/primary/p/224664/.

[mooc] Start des MOOCs „Open Educational Resources in Higher Education“ #imoox #unite! #tugraz

Es freut uns sehr, dass nach langer Vorbereitung es nun endlich soweit ist – der MOOC zu „Open Educational Resources in Higher Education“ startet und es ist ein MOOC mit vielen Besonderheiten, denn das Ziel ist hiermit eine Lehrveranstaltung im Rahmen der europäischen Allianz UNITE! anzubieten. Damit stehen wir unweigerlich vor der schwierigen Herausforderung verschiedenster Landessprachen, wenn wir nicht uns generell auf eine verständigen. Im Zeitalter künstlicher Intelligenz haben wir daher uns zu einem Experiment entschlossen – wir wollten die Videos in allen Landessprachen anbieten, vorgetragen von KI-generierten Avataren und unterstützt von KI-Software was die Übersetzungen betrifft. Wie gut uns das gelungen ist, kann man nun einsehen – also anmelden und mitmachen, wir hoffen ja auch, dass der Inhalt ansprechend ist.

Open Educational Resources (kurz OER) sind Bildungsmaterialien, die frei zugänglich sind und bearbeitet, angepasst und weitergegeben werden können, ohne gegen Urheberrechte zu verstoßen. Das gelingt, weil dazu sog. „offene Lizenzen“ oder gemeinfreie Materialien genutzt werden.
In diesem Kurs wird in das Konzept von Open Educational Resources für Studierende und Lehrende an Hochschulen eingeführt. OER sind nicht nur für internationale Vereinigungen wie UNESCO ein wichtiges bildungspolitisches Thema, OER sind auch ein Bestandteil der Forderung nach „Open Science“.
Im Kurs wird zudem gezeigt, wie und wo OER gesucht und veröffentlicht werden können. Und natürlich werden auch offene Lizenzen wie CC-0, CC BY und CC BY-SA erklärt.
Hintergrund dieses mehrsprachigen MOOC-Angebots ist das gemeinsame Projekt „Unite! OER courses“ aller neun Hochschulen der europäischen Universitätsallianz Unite!.

Das erste Video im MOOC (in der Sprache deutsch):

YouTube

Mit dem Laden des Videos akzeptieren Sie die Datenschutzerklärung von YouTube.
Mehr erfahren

Video laden

[Link zur kostenlosen Anmeldung]

WICHTIG: wenn man eine andere Sprache ausprobieren möchte, einfach im Menü die entsprechende wählen, die Videos werden dann automatisch in dieser Sprache zur Verfügung gestellt.

Und nein, weder Sandra noch ich sind einem dieser Videos zu sehen, es sind ausschließlich unsere Avatare 🙂

[video] XBox Natal Projext

Ein sehr interessantes Projekt der Spieleindustrie ist XBox Natal – man spielt einfach durch Gesten und braucht kein Steuergerät mehr. Wenn man sich das durch den Kopf gehen lässt, dann kann dies durchaus auch für die Lehre interessant sein: Was ist wenn man Präsentationen einfach durch Bewegungen steuern bzw. Simulationen und Experimente durchführen könnte um damit den Unterricht erlebbarer und angreifbarer zu machen? Nun ich glaub, da ist viel Potential vorhanden und man sollte diese Gedanken weiterverfolgen. Jetzt aber das Video:

Weiterer Bericht: How I Became An Avatar