The XAI guide addresses different aspects of explainable AI in the individual chapters, so that the guide serves as an introductory or in-depth guide for people with little prior knowledge as well as for people with experience. The XAI guide is therefore aimed at domain experts and end users such as employees and managers.

AI methods offer the possibility of finding solutions for complex classification tasks or the identification of patterns in large amounts of data. Depending on the AI model used, the way in which AI methods make their decisions is often incomprehensible to humans. This is where the research field of XAI comes in, which investigates how algorithmic decisions made by AI systems can be explained in a way that humans can understand. XAI methods pursue one of two approaches: Either they make the AI model and the associated decision-making process directly comprehensible or they explain a decision retrospectively ("post hoc"). The XAI guide offers short texts on important XAI aspects. It contains a description of key AI and XAI terms as well as individual XAI procedures, the benefits, target groups and challenges of XAI and the consideration of XAI in EU regulations (GDPR, draft AI regulation). The individual chapters build on each other, but are also largely understandable independently of each other. The guide thus aims to impart knowledge in the field of XAI.

The XAI guide can be downloaded from the website as a PDF document. Supplementary offers are available in the form of the XAI canvas and XAI assistant, which offer a user-centered approach to selecting an XAI method, as well as in the form of an XAI workshop.

Not necessary.