XAI4Covid is as part of the supplementary funding module of the Volkswagen Foundation that aims at supporting research projects whose findings not only contribute directly to overcoming the crisis, but can also provide impetus for overcoming major societal challenges in the medium to long term. XAI4Covid investigates how research on explainability of AI systems could be translated and applied to improving public communication in complex crisis situations such as the COVID-19 pandemic. We will examine which AI explanation techniques could be applied to increase the understandability of public communication of expert knowledge (e.g contrastive and counterfactual explanations) and how they should be adapted to increase public understanding of expert decisions in crisis situations (e.g. by integrating persuasive communication techniques such as storytelling). This shall provide experts and decision-makers with novel methods to improve their communication and increase public acceptance and implementation of their recommendations (e.g. COVID-19 containment measures, vaccination).
To this end, we have developed, among other things, an interactive tool that allows the general public to easily explore indoor infection risks based on various influencing factors (masks, incidence, ventilation):
The interactive simulator is based on scientific models and has been scientifically evaluated for user acceptance and effectiveness.
A summary of the evaluation can be accessed here, a publication with all results is currently in preparation. To find out how AI research may strengthen your arguments, please have a look at our Guidelines on Explanation Methods for COVID-19 Communication (in German).
Furthermore, if you you want to learn how you can easily create an interactive teaser story for your Instagram feed for the simulator, this tutorial (in German) will take you through the process and provide you with the video footage you need.
Our project poster provides futher information about our interdisciplinary approach to human-centered COVID-19 communication and how we applied insights from Explainable AI research.
- Date Friday, October 22nd, 2021
- Tags Social Innovation, Active, Participatory Systems