CAIS Symposium – AI in Education in Germany: Current State and Future Developments
The symposium aims to facilitate discussions on the current state of AI in education research and implementation in Germany, while also highlighting key issues, challenges, and directions for future developments.
Review: CAIS Symposium
On June 29 and 30, the CAIS research program „Educational Technologies and Artificial Intelligence“ hosted the symposium „AI in Education in Germany: Current State and Future Developments“. The symposium took place at Beckmanns Hof (Ruhr-Universität Bochum) and brought together experts working on research and implementation of educational technology.
The symposium started with welcoming remarks by Nikol Rummel, head of the research program. It continued with a tour of a poster gallery, where participants had the opportunity to get to know each other and discuss their research foci and interests in a friendly and productive atmosphere.
Three keynotes provided inspiration and food for thought. In her talk „Towards hybrid human-AI learning technologies“, Inge Molenaar (Radboud University, NL) discussed levels of automation and the potential roles of AI, learners, and teachers. Vincent Aleven (Carnegie Mellon University, US) shared his vision of the ways AI can enrich and facilitate learning in his talk „Harnessing human-AI synergy in the smart classroom: What works, new opportunities“. Finally, Wayne Holmes (UCL, UK) provided critical insight and drew the audience’s attention to the ethical issues of using AI in educational contexts („The Ethics of Artificial Intelligence and Education. A critical studies approach“).
An interactive format was chosen for the other sessions in the symposium. Three panel discussions were held. Steffen Albrecht (Office of Technology Assessment at the German Bundestag), Wayne Holmes, Inge Molenaar, and Farina Nagel (RuhrFutur) discussed their perspectives on the challenges of using AI in education. In the second session, Vincent Aleven, Frank Reinhold (Freiburg University of Education), and Torsten Zesch (FernUniversität in Hagen) shared their experience with design and evaluation of AI systems for education. The third panel discussion featured Iris Backfisch (Eberhard Karls Universität Tübingen), Elisabeth Bauer (Technische Universität München), and Timo Leuders (Freiburg University of Education), who exchanged ideas on empowering educators to harness the power of “intelligent” tools: professional training of teachers, their needs and competences, and readiness of the educational system for AI-based tools. Both days of the symposium ended with group works to revisit some valuable points from the previous sessions, discuss controversies and challenges, and determine the most urgent and burning questions when it comes to developing AI in education in Germany.
Keynotes
Towards hybrid human-AI learning technologies
There are multiple scenarios in which artificial intelligence (AI) could improve teaching and learning. In this talk, the augmentation perspective and the concept of hybrid intelligence are introduced to frame our thinking about AI in education. The involvement of different stakeholders (i.e., researchers, education professionals, entrepreneurs, and policymakers) is necessary to understand the opportunities and challenges of AI in education. In order to facilitate a meaningful dialogue, a common language about AI in education is introduced. The six levels of automation model develop our thinking about the roles of AI, learners, and teachers. In this model, the transition of control between teacher and technology is articulated at different levels which are illustrated with different examples. Next the detect-diagnose-act framework is used to describe the core functions of AI in education and to discuss the future of AI in education with different inspiring innovative cases. This talk will elaborate on Hybrid Human-AI learning technologies that augment human intelligence with artificial intelligence to empower learners and teachers in their quest to make education more efficient, effective and responsive.
Bio Inge Molenaar
Inge Molenaar is the director of the National Education Lab AI (NOLAI) and associate professor of educational sciences at the Behavioural Science Institute at Radboud University in the Netherlands. She has over 20 years of experience in technology-enhanced learning, taking multiple roles from entrepreneur to academic. Her research focuses on technology-empowered innovations to optimize human learning and teaching. The application of data, Learning Analytics and Artificial Intelligence in understanding how learning unfolds over time is central to her work. Artificial Intelligence offers a powerful way to measure, understand and design innovative learning scenarios. Dr Molenaar envisions Hybrid Human-AI Learning Technologies that augment human intelligence with artificial intelligence to empower learners and teachers in their quest to make education more efficient, effective and responsive. In the National Education Lab AI (NOLAI), a collaboration between the Dutch government, schools, research and industry, we believe that a multistakeholder, interdisciplinary approach is essential to develop the next generation educational technologies and a shared understanding of responsible use of AI in education. Dr Molenaar has received multiple grants among which an ERC Starting Grant to develop the first Hybrid Human-AI Regulation system to train young learners’ Self-regulated learning skills with AI and she is a Jacobs Foundation Fellow. She holds Master’s degrees in Cognitive Psychology and International Business studies and a PhD in Educational Sciences (University of Amsterdam).
Harnessing human-AI synergy in the smart classroom: What works, new opportunities
What should the smart, AI-supported, classroom look like? We should of course continue to use AI in ways that have proven to be effective, such as in supporting deliberate practice with AI-based tutoring software. Yet even with the expanding use of AI-based tutoring systems, difficult and enduring challenges in the educational system remain, such as large opportunity gaps between different population segments. To create a smarter classroom, it will help to redesign the role of humans and AI in the educational system, enhancing both, in a way that is maximally synergistic. AI could support more humans in a wider range of roles as they facilitate learning: students, teachers, tutors, and parents could all, with the help of AI, assist students. Some of these forms of AI-based support have been shown to be successful in research studies and are beginning to transition into educational practice. Other forms are just beginning to be addressed by researchers. Human-centered design will be key to harnessing human-AI synergy. I will illustrate progress and opportunities with examples from my own, and others’ research.
Bio Vincent Aleven
Vincent Aleven is a Professor in the Human-Computer Interaction Institute at Carnegie Mellon University. He has 25 years of experience in research and development of AI-based learning technologies grounded in cognitive theory, self-regulated learning theory, and user-centered design. His lab created the Mathtutor website with AI-based tutoring software for middle school mathematics. They also created CTAT and Tutorshop, tools for the development and deployment of AI-based tutoring software. Aleven has over 250 publications to his name. He and his colleagues and students have won 11 best paper awards at international conferences. He is or has been PI or co-PI on over 20 major research grants. He is co-editor-in-chief of the International Journal of Artificial Intelligence in Education (IJAIED). Aleven is a co-founder of the Pittsburgh-based company Carnegie Learning, which markets mathematics curricula for middle school and high school that include AI-based tutoring software, a central focus of his research.
The Ethics of Artificial Intelligence and Education. A critical studies approach.
Artificial Intelligence (AI) is frequently hailed as a ‘solution’ to many of education’s core problems (e.g., OECD, 2021) – problems such as student underachievement, assessment at scale, and better preparing learners for 21st century career paths. However, such claims tend to be aspirational rather than evidence-based (Miao & Holmes, 2021), and overly simplistic, forgetting issues such as agency, pedagogy, surveillance, efficacy, and ethics (Holmes et al., 2021; Holmes et al., 2022; Holmes & Porayska-Pomsta, 2022; Porayska-Pomsta, Holmes and Nemorin, 2023). In fact, current applications of AI in Higher Education tend to be solutions- rather than problems-oriented, and all too often undermine student agency and disempower educators; while the teaching of AI almost always focuses on the technological dimension of AI to the exclusion of the human dimension (its ethical, human, and social justice implications). Accordingly, this presentation will explore the application and teaching of Artificial Intelligence in education from a critical studies and human rights perspective. It will identify and address many of the key myths, it will argue for an ethics by design approach, and it will pose more questions about AI and the futures of learning than it answers.
Bio Wayne Holmes
Wayne Holmes (PhD, University of Oxford) is an Associate Professor in the IOE, UCL’s Faculty of Education and Society, at University College London. His research takes a critical studies perspective to the teaching and application of Artificial Intelligence in educational contexts (AI&ED), and their ethical, human, and social justice implications. Wayne is leading the Council of Europe’s AI&ED project, which is developing a legal instrument to regulate the application of AI in education (AIED) and a recommendation for teaching about AI (AI Literacy), and for which he led the report: Artificial Intelligence and Education. A critical view through the Lens of Human Rights, Democracy and the Rule of Law. He is also Consultant for the Technology and AI in Education unit at UNESCO, for which he co-wrote AI and Education: Guidance for Policy-makers, and he is Senior Researcher in AI&ED for IRCAI (the International Research Centre on Artificial Intelligence under the auspices of UNESCO). Wayne has also co-written Artificial Intelligence in Education. Promise and Implications for Teaching and Learning (Holmes et al., 2019), Citizens Interacting with AI Systems (for the EU JRC, Vuorikari and Holmes, 2022), State of the Art and Practice in AI in Education (Holmes and Tuomi, 2022), and The Ethics of AI in Education. Practices, Challenges and Debates (Holmes and Porayska-Pomsta, Eds, 2022). He has also edited a Special Issue (in press) of the International Journal of Artificial Intelligence in Education, entitled ‘Artificial Intelligence in Education. Coming of Age?’. He has also given invited keynotes about AI&ED in countries around the world.
CAIS Symposium
Inge Molenaar
Radboud University, NL
Towards hybrid human-AI learning technologies
Vincent Aleven
Carnegie Mellon University, US
Harnessing human-AI synergy in the smart classroom:
What works, new opportunities
Wayne Holmes
University College London, UK
The Ethics of Artificial Intelligence and Education. A critical studies approach.
Panels
Invited panel speakers will engage in a comprehensive discussion on the critical questions surrounding the research and design of AI systems for learning, while also delving into the broader contexts of preparing for and regulating the implementation of AI-based tools in education.
Challenges of using AI in education
When considering the integration of „intelligent“ systems into educational settings, it is imperative to thoroughly address the ethical and regulatory challenges associated with their implementation. In this panel, we will discuss the following questions: What are the potential risks and implications that arise from utilizing these systems? Who are the various stakeholders involved in the implementation process? Besides incorporating their interests, how can we ensure a holistic and inclusive approach to integrating AI for learning? Finally, how can we examine and determine the potential societal implications that may emerge from the widespread adoption of AI in education?
Design and evaluation of AI systems for education
Understanding the current research practices in designing „intelligent“ systems for educational settings is key to their effective implementation. In this panel discussion, we will explore the learning goals for which these systems can be employed and identify the variables that may be influenced by their implementation. Moreover, we will discuss various dimensions that are important when assessing the effectiveness of these systems ensuring a comprehensive evaluation.
Empowering Educators to harness the power of “intelligent tools”
In this panel discussion, our primary focus is to gain valuable insights into the perspectives of educators from diverse learning settings. We explore the essential requirements for equipping teachers and lecturers for successful AI integration while catering to their specific needs. We shed light on the crucial competencies that educators must possess to effectively utilize „intelligent“ tools and address the critical issue of bridging the gap between evidence-centered design and the practical implementation of AI in education.