KonCheck - Combating misinformation using AI
Our research group, in collaboration with our project partners, is tackling this challenge by exploring how AI can help citizens stay politically informed without falling prey to fake news and disinformation. Funded by the Daimler and Benz Foundation with €1.5 million over three years under the “Ladenburger Kolleg” program, our project aims to create an innovative, user-friendly app and explore the effects of using this app on political opinions and information literacy.
KonCheck - the AI tool we will be developing in this project - will contextualize politically relevant information and verify its authenticity using Large Language Models (LLMs), commonly referred to as AI. Users can ask questions about specific texts, view sources, access articles in simple language, and understand content in context. Designed for intuitive use, the tool targets vulnerable groups such as people with cognitive impairments, seniors, and young first-time voters.
Our goal is to foster a new form of participation in the digital world. Instead of merely consuming media content, KonCheck will enable users to interactively engage with it.
Our interdisciplinary team includes the Social Research Centre Dortmund (Prof. Jürgen Howaldt / Prof. Johannes Weyer) and the Department of Communication for Health, Inclusion, and Participation (Prof. Matthias R. Hastall) at TU Dortmund, as well as the CISPA Helmholtz Center for Information Security, the Institute for Internet Security at the Westphalian University of Applied Sciences, and the Rhine-Waal University of Applied Sciences.
Official KonCheck Press Release of the TU Dortmund: https://www.tu-dortmund.de/detail/neuer-forschungsverbund-zu-ki-und-demokratie-erhaelt-15-millionen-euro-44085/
Official KonCheck Press Release of the Daimler & Benz Foundation: https://www.daimler-benz-stiftung.de/cms/de/forschen/ladenburger-kollegs/neuer-forschungsverbund-ki-basierte-methoden-zur-unterstuetzung-von-meinungsbildung-und-partizipation.html