Box 19 – Explainable AI

Box 19: Explainable AI

Trust, but verify!

The functioning of AI systems that are trained on large amounts of data using machine learning algorithms is often difficult to explain. Accordingly, it is difficult to determine how such an AI system has come to a result and how to assess it. AI systems that apply methods of explainable artificial intelligence (eXplainable AI) can help with this. In this box, students test how their confidence in the results of AI systems changes as a result of explanations for how the results were obtained.

The information text explains how eXplainable AI methods work.

This learning box was developed in cooperation with Ute Schmid, Chair of Cognitive Systems at the Otto-Friedrich-University Bamberg. Many thanks for your cooperation!