As part of a statewide pilot project in Baden-Württemberg, the use of generative AI (GenAI) in schools is being tested for the first time. In close cooperation with the Center for School Quality and Teacher Education (ZSL), we are conducting an accompanying research study within a consortium consisting of the Hector Institute for Empirical Educational Research, the Leibniz-Institut für Wissensmedien, and the Tübingen Center for Digital Education. The GenAI application fAIrchat, integrated into Moodle, serves as a GDPR-compliant AI-based learning assistant for both students and teachers.

The aim of the research is to systematically model the practices of GenAI use by students and teachers. We analyze how GenAI is integrated into classroom instruction, how it is accepted, and under which conditions it effectively supports learning. The theoretical foundation is based on technology acceptance models, complemented by an analysis of so-called boundary conditions, the contextual factors that enable responsible and effective use of GenAI in schools.

The project provides the first empirical insights into the design of AI-supported learning and contributes to the evidence-based development of digital education in authentic school contexts.