Human-centered approaches in working environments needs ethical reflection—Experiences from KARL

Abstract In our paper, we discuss some insights from our work in the regional Centre of Competence of Artificial Intelligence (KARL), which deals with the implementation and application of Artificial Intelligence in working and learning environments. We present a conceptual framework that we have developed to define social, legal and ethical dimensions of AI in working environments. As a first step, the article describes the definition of these aspects within the technical development processes in KARL with an emphasis on ethical issues. Furthermore, a practical example is used to illustrate how normative premises can be elaborated to inform the design process of AI-systems. Next, the process of development and implementation of AI-systems in working environments is taken into account. The article explicitly emphasises the importance of ethical reflection, i.e. on norms such as fairness, social sustainability and the creation of meaningful work, to inform in human-centered approaches to AI-based work. Practical Relevance : This paper discusses a process-oriented, human-centred approach to AI-based work. The joint negotiation and definition of “human-centred work with AI” in each specific context with all concerned stakeholders lies at its heart. The concretisation of these implications is of great importance for the practical implementation of the overarching concept, as it is only in practice that it becomes clear whether and how a participatory design approach is successful, which areas of tension and problems arise and how these can be dealt with.