An application-oriented research project at the intersection of social impact and Explainable AI has officially launched: the BMFTR-funded project EKILED (Erklärbare KI zum Lernen über digitale Desinformation) aims to make AI systems for detecting disinformation more transparent and educationally valuable.

Against the backdrop of growing challenges posed by digital disinformation in today’s interconnected public sphere, EKILED focuses on advancing Explainable AI approaches to open up the “black box” of complex AI models. The goal is to help users—particularly students—understand how and why AI systems classify content as disinformation, and which patterns they rely on in doing so.

At the core of the project is the development of an interactive demonstrator designed to act as a reflective learning partner. This tool will support learners in critically assessing digital content and independently identifying disinformation. At the same time, it aims to provide teachers with an accessible and practical way to foster media literacy and critical thinking in the classroom. The value of the interactive demonstrator will be evaluated in the field in collaboration with schools.

The collaborative project brings together the WIN research group Information Systems IV - Digital Platforms & Services, the University of Ulm, the University of Bamberg, DASU – Transfer Center for Digitalization, Analytics & Data Science Ulm, and Eduversum GmbH.

Funding is provided by the Federal Ministry for Research, Technology and Space (BMFTR) for the period from April 2026 to March 2029.