Promoting Human Competences by Appropriate Modes of Interaction for Human-Centered-AI

Abstract

There is an ongoing discussion about human-centered AI (HCAI) that emphasizes the value of including humans in the loop. We focus on types of HCAI in the context of machine learning that synergistically combine the complementary strengths of humans and AI and seek to develop competencies and capabilities of both parts. The development of human competencies is a largely neglected aspect compared to criteria such as fairness, trust, or accountability. Based on early discussions about the role of humans in the use of expert systems, the current HCAI discourse, and a literature review, we identify 10 modes of interaction that represent a way of interacting with AI that has the potential to support the development of human competencies relevant to the domain itself, but also to its context and to the use of technologies.