EdReNe 21st Strategic Seminar on AI in Education

The objective of the Educational Repositories Network (EdReNe) meetings is to nurture forums for the exchange of expertise and best practices in managing educational content repositories at the K-12 level in Europe. The 21st EdReNe Strategic Seminar was held in Brussels on 30-31 May 2024, and brought together key stakeholders in education and technology for a focused two-day forum on artificial intelligence in education. 

Organised by KlasCement and European Schoolnet, the event was attended by 26 representatives of Ministries of Education and other educational authorities and institutions from 10 European countries (Belgium, England, Germany, Greece, Ireland, Italy, Latvia, Norway, Slovenia, Türkeye).

The seminar featured a mix of presentations and group discussions centred around the responsible deployment of Artificial Intelligence in education. Participants shared ideas and expertise on strategies for balancing the use of new technological solutions for educational purposes while mitigating any potential negative impact on teaching and learning. A recurring theme in all presentations and discussions emphasised the significance of providing targeted training and support for teachers and students interested in utilising artificial intelligence within educational settings.

Belgium’s Kenniscentrum/Knowledge Center Digisprong presented their Vision Paper for the “Responsible Use of AI in Education”. The paper is the result of a collaborative process that has involved teachers, students and other school stakeholders in Belgium Flanders - including trading associations, students and parents’ associations, union of ICT coordinators, educational and AI experts - in a broad discussion about the impact of AI in schools.

Given the assumption that AI is here to stay, the vision paper seeks to assist learners, educational professionals, and organisations in Belgium Flanders. To this purpose, it begins by defining the concept of responsible AI in education and subsequently highlights the essential conditions for teachers and students to use AI responsibly, adhering to seven fundamental ethical requirements:

  1. The learning process of the learner is paramount from a pedagogical-didactical and socio-emotional perspective. Everyone involved in the learning process has an important role to play in interaction with one another.
  2. AI is not an end in itself. It is a possible means for achieving educational goals and must have added value for education. 
  3. AI applications in education are trustworthy. Human autonomy, diversity, transparency, etc.  must be respected.
  4. AI applications in education are based on shared values. The use of AI-applications has to be in line with the values that the school has prioritised. 
  5. Responsible AI is a continuous process. Choosing AI is not a one-time exercise. It is an ongoing process of responsible development, procurement, use and evaluation.
  6. Education has a support network that is AI-ready and AI-resilient. Schools can rely on the extended network. Teachers know who to consult with in case of issues or concerns.
  7. Professionalisation and responsible AI go hand in hand. Digital literacy is crucial.

To make the above principles tangible and inform the process of teaching and learning in schools, the group has developed a series of “Guidelines for Schools” which include the following key actions to ensure responsible use of AI in education:

  1. Define what you want to achieve.
  2. Ask yourself whether you should achieve this with AI or if other means are better suited.
  3. Apply moral values, ethical framework and regulations.
  4. Work together with all the parties involved.
  5. Take joint responsibility.
  6. Assess your AI applications regularly.

During the subsequent World Café sessions, participants delved into the seven fundamental requirements for AI in Flemish Education, exchanging insights on key initiatives within their countries and organisations aimed at integrating AI into educational platforms. A summary of the key findings from these group discussions is accessible here.

The seminar continued with presentations showcasing international initiatives related to the integration of artificial intelligence in education.

Prof Şahi̇N İdi̇l from The Scientific and Technological Research Council of Türkiye (TÜBİTAK) illustrated the national strategies and models for the implementation of artificial intelligence in Turkey. These are based on six main pillars:

  • Training AI experts
  • Supporting research entrepreneurship and innovation
  • Expanding access to quality data and technical infrastructure
  • Implementing regulations to accelerate socio-economic adaptation
  • Strengthening international collaborations
  • Accelerating structural update, etc.

As part of the middle and high school curriculum, a dedicated 8-week course has been designed to teach students ethical AI usage. Examples of content and models used in this AI training for students were presented, including practical examples that highlight the advantages of integrating AI into education systems and the skills students can develop through personalised and self-regulated learning.  

Sigurd Trageton from the Norwegian Digital Learning Arena (NDLA) explained how AI applications have been used in Norwegian upper secondary education with the aim to provide: 1) secure access to generative AI for students; 2) explore possibilities of AI to produce learning resources; 3) develop AI integrated functions to support teachers and students and to get feedback.

Patrick Coffey, National Coordinator for Scoilnet & Digital Content (Ireland) presented the outcomes of the Erasmus+ funded project entitled ‘Artificial Intelligence for and by Teachers’. AI4T was a three-year project which explored the incorporation of artificial intelligence into educational experiences at primary and post-primary levels and looked into the potential implications for teachers. The main project output was a MOOC in 5 languages complemented by a textbook for training teachers.  

Dr Susanne Friz from the FWU Institut für Film und Bild in Wissenschaft und Unterricht gGmbH presented Germany’s Adaptive Intelligent System (AIS) project to support teachers and learners of all school levels. 

The project, initiated in March 2024 and spanning 28 months, receives funding from the German Federal Ministry of Education and Research through the Digital Pact for Schools. All 16 states in Germany are participating, with the FWU leading the implementation.

The project aims to create an AI-supported adaptive computer learning environment called AIS. This system will significantly contribute to individual student support and teacher professionalisation. By flexibly integrating into teaching phases, AIS will establish a framework for modern, varied, motivating, and adaptive teaching and learning. Key features include personalised learning paths based on each student’s level, powered by an intelligent algorithm. The learning environment will offer direct feedback on student progress, leveraging developments in LLM (Learning and Learning Management) and AI. The target system, set to launch in 2026, will provide adaptive learning content and exercises, supporting teachers in the classroom and in fulfilling their tasks and empowering students to work independently at their own pace, both in class and at home. Two subject examples — one STEM and one from the humanities—demonstrated the system’s potential during the presentation. 

Wendy DeKeyser from KlasCement (Blegium-Flanders) presented the new MaaKC project, which focuses on creating interactive lesson materials directly on the KlasCement platform. During the Covid-19 pandemic, many teachers recognised the need for an authoring tool directly integrated into the platform. The potential functionalities of the tool in development were brainstormed over a five-day process using a design strip.

The system will empower teachers by allowing them to search for relevant resources based on keywords. They will then be able to select a resource and enhance its interactivity by adding exercises, questions, and other modifications. Teachers will also be able to test the resource to ensure mobile friendliness, collaborate with colleagues during preparation, and save the adapted content to their personal space. Additionally, they will have the option to share these resources within the KlasCement community. Whether starting from scratch, using a template, or requesting AI-generated content, teachers will have flexible options for creating engaging educational materials.

Finally, Knut Inge Skifjeld from the Norwegian Digital Learning Arena (NDLA) presented Norway’s platform for H5P content administration. In a world that grows increasingly complex, creating and sharing content remains paramount. It is therefore time to transition from ‘Nice.net’ to ‘Smart.net.’ To achieve this, Digisprung and NDLA have been exploring collaboration opportunities centred around a specific open-source software for serving H5P content across various systems and platforms. NDLA’s focus lies in developing open-source content creation tools, accessible to a global audience through well-documented and user-friendly packaging. Cooperation and reuse are essential keywords for achieving the necessary quality in OER tool development. Additionally, the integration of generative technology and machine translation into H5P content types presents intriguing possibilities.

The EdReNe 21st Strategic Seminar concluded by emphasising the significance of international knowledge exchange within the EdReNe network. Participants appreciated the chance to meet face-to-face and expressed their desire to enhance information sharing and collaboration opportunities in projects related to integrating AI into educational platform solutions and other initiatives.

You can access the presentations and resources shared during the 21st Educational Repositories Strategic Seminar below: