by Peter Andries, Media & Learning Association, Belgium.
With the recent release of ChatGPT based on GPT-3.5 and its wide-open access, interest in AI has seen a massive surge now that a broad adoption of AI seems closer than ever. This prospect not only raises questions about how AI can be used to best facilitate education, but it also makes us think further about its impact on learning pathways, ethics, IPR, media literacy and student cheating, just to name a few concerns.
The European Digital Education Hub, an open online collaborative community for digital education stakeholders in Europe, recently organised an online event to offer an introduction on AI in education, entitled “The Myths and the Potential of AI in Education”. While the focus of this event was not specifically aimed at higher education, it did address applicable principles, developments, and perspectives very much relevant to our Media & Learning community.
Maria Gkountouma of the European Commission DG EAC gave a summary of what has been put forward in the Ethical guidelines on the use of artificial intelligence and data in teaching and learning for Educators as launched in October 2022. It is available in 24 languages and written to accommodate educators from all backgrounds to understand what can be expected of AI to support teaching, learning and assessment while at the same time making sure that both practical and ethical considerations are being covered. The document does this by offering clear use cases and examples to make you reflect on the implications of incorporating AI in education.
The next talk was given by Tomas Krilavičius, Dean of Faculty of Informatics, Vytautas Magnus University, and he presented an AI tool developed to identify hate speech. The term “hate speech” is used when negative attributes are assigned to people because they are a member of a particular group. The AI tool classifies speech (e.g., social media post, manifests, etc.) by sorting it into three categories: hate, offensive and neutral speech together with a probability score. While currently focused on Lithuanian, with some more development and open access, an AI programme like this can be a very useful tool to teach certain aspects of media literacy or to automate content moderation.
Thierry de Vulpillières, CEO & co-founder EvidenceB, provided an introduction and showcase of an adaptive learning platform that uses AI to support K12 students and teachers to offer differentiation and personalisation while teaching maths. The platform profiles the students to assess their knowledge and implements reinforcement learning to optimally challenge each student to avoid the chance of drop out due to lack of understanding or attention loss. It also provides the teacher with a simple dashboard to follow the students’ progress and challenges to be able to provide targeted lessons to individuals or groups.
The final presentation on the Teachers’ perspective on the potential and risks of AI use in education was given by Lidija Kralj, EduConLK, who was also chairing the session. She addressed the myths, potential, promises, and ethical questions implementing AI in education. She also presented the input from the wider community of practice of the European Digital Education Hub discussing the challenges that teachers are facing to acquire and develop the necessary skill set for AI-enabled education. What competences do you need and what should be included in the practical toolkit for active in-class teachers to become quickly operational in using AI? If you wish to learn more about AI in education, make sure to sign up for this newsletter as well as to register on the European Digital Education Hub to become part of the community of practice and gain access to its upcoming events, active discussions and resources.
Author
Peter Andries, Media & Learning Association, Belgium
