The project Bridge is co-funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the Servicio Espanol para la Internacionalizacion de la Educacion (SEPIE). Neither the European Union nor the National Agency SEPIE can be held responsible for them. -- Number: 2025-1-ES01-KA220-HED-000359236 --
The BRIDGE Digital Museum brings together good practices, educational proposals and images that inspire reflection and action
Social Challenges
Analyse social challenges to intercultural dialogue, social inclusion and democratic cohesion
Good Practices
Discover experiences that generate social impact
Educational Space
Access teaching guides, activities and materials for the classroom
Documentary Photography
Explore the social challenges in a visual, reflective and educational way
BRIDGE Main Challenge: A.7. Digital ecosystem, disinformation and the need for media literacy and critical thinking
Generative AI and algorithmic decision-making systems increasingly shape how citizens form opinions. Hallucinations, model collapse and biased training data spread unreliable content and reproduce discrimination, while the technology is presented as neutral. This threatens the epistemic agency and independent political judgement that democracy depends on, and educators lack tools to address it.
AI.D offers a modular Critical AI Literacy framework for higher education, VET and non-formal education: three modules and ten teaching units with trainer guides, plus active methods such as AI.Dialogue, the Me & AI barometer and the AI Pub Quiz. Learners analyse AI as a political project, identify bias and disinformation and co-create institutional rules for its use.
AI.D (Artificial Intelligence and the Shaping of Democracy) is a transnational project led by Leibniz Universität Hannover with partners in Austria and Belgium. It develops a Critical AI Literacy framework for higher education, vocational training and non-formal education. Through three modules, ten teaching units and methods such as AI.Dialogue, students and teachers learn to recognise disinformation and algorithmic bias, to question claims of technological neutrality and to co-create democratic rules for the use of AI in their institutions.
AI systems and large language models have become gatekeepers of public opinion. Unreliable outputs, deepfakes and biased training data amplify disinformation and existing discrimination, for instance by rendering disabled or non-binary people invisible or penalising diverse CVs. Citizens risk losing the independent political judgement that democratic participation requires.
In Germany polarising discourse is becoming normalised and the digital environment amplifies hate and disinformation. Higher education has a strategic role in building critical digital literacy. AI.D moves away from technical instruction towards questions of power, equity and democracy.
AI.D responds directly to Challenge A7 by extending media literacy to Critical AI Literacy: learners identify hallucinations, bias and synthetic content, analyse who profits from AI and deliberate on its governance. It also feeds Challenge A6 by modelling bottom-up decision-making in institutions.
A modular system of three thematic blocks (fundamentals, systemic implications, governance and democracy) with ten teaching units and trainer guides. Active methods include the AI.Dialogue role-play with three assigned positions, the Me & AI barometer, discourse and SWOT analysis, quizzes and prompt exercises. All materials are bilingual OER that teachers can download, edit and adapt.
AI.D training modules: three modules and ten teaching units with trainer guides
Educational resource
The core of the practice is a modular curriculum organised in three thematic blocks: (1) Fundamentals and Reality, which demystifies AI and distinguishes rule-based automation from machine learning; (2) Systemic Implications, covering AI in everyday life, filter bubbles on social media and AI tutors in education; (3) Governance and Democracy, examining the EU AI Act and AI as a political and ideological project. The ten teaching units add up to about 750 minutes of instruction. Each unit comes with a detailed trainer guide stating learning objectives, estimated timings and the sequence of activities, together with interactive slide decks and ready-to-use handouts on automation risks and algorithmic ethics. The modules are aligned with DigComp 2.2 and the UNESCO competency framework for teachers, and are available in English and German under CC BY-SA 4.0.
Teaching staff, Students
AI.Dialogue: role-play deliberation method on AI governance
Manual
AI.Dialogue is the project's main participatory method. Participants are assigned one of three controversial positions towards AI (anxious, sceptical or optimistic) and deliberate about how AI should be governed in their own institution or community. The exercise is designed to make the range of views in a group visible, to expose personal assumptions and to lead from abstract debate to concrete proposals for self-governance. In the governance strand of the modules it is used to co-create 'institutional compasses': criteria for the ethical use of AI in teaching and learning that teachers and students negotiate together. A manual with instructions is part of the OER package, so the method can be run in higher education, vocational training and non-formal settings.
Teaching staff, Students, Professionals
https://aidstorage.blob.core.windows.net/aid/modules/AID_DM_AIDialogue_EN.pdf
Me & AI positions barometer and 15 Minutes News methodology
Educational resource
Two short activating formats complement the longer teaching units. The Me & AI barometer is a physical positioning exercise: participants move and position themselves in the room in response to statements about AI, which makes the spectrum of opinions in a group visible and prompts reflection on personal biases. 15 Minutes News is a compact methodology for working with current news about AI in a short time slot. Both are active methodologies with their own instruction manuals and can be used independently of the full modules. Teachers who want to introduce critical AI literacy in an existing course can therefore start with either of them.
Teaching staff, Students
AI Pub Quiz, matching games and Mega Prompts exercises
Interactive game
A set of playful tools supports the modules. The AI Pub Quiz tests and consolidates knowledge about how AI systems work and where they fail, in a format suited to larger groups and non-formal settings. Matching games let learners practise the concepts of the modules, for example the difference between rule-based automation and machine learning. The Mega Prompts exercises are practical prompt-engineering tasks in which participants experiment with language models and observe how outputs depend on framing, where hallucinations arise and where bias appears. The tools are intended to demystify the technology through hands-on experience and to give learners material for the subsequent critical discussion. They are part of the OER package.
Teaching staff, Students, General Public
Explanatory videos and webinars on AI hallucinations and model collapse
Educational videos
Short explanatory videos and recorded webinars present two technical phenomena in a form suited to classroom use. 'AI hallucinations' refers to confident but false outputs of language models; the project links this to the EBU/BBC 2025 finding that 45 per cent of AI-generated answers on current news contained significant problems of accuracy or reliability. 'Model collapse' describes the degradation of quality when models are trained recursively on synthetic data. The videos can be used as preparation before a session or as input within the teaching units. They are part of the transmedia materials on the project's bilingual platform, which is intended to continue offering webinars and updated materials after the project ends.
Teaching staff, Students, General Public