Language development is one of the key challenges in early childhood education. Despite intensive efforts by the federal states, accompanying studies on the effectiveness of support measures regularly yield sobering findings (Schneider, 2018), with rates of children requiring support ranging from 15% to 41% depending on the state (Autorengruppe Bildungsberichterstattung, 2020).

One empirically proven method is dialogic reading, which is based on facilitating language-promoting interactions during read-aloud sessions (Ennemoser et al., 2013; Mol et al., 2008). Widespread adoption in early childhood education settings has failed in traditional implementations due to factors such as a lack of personnel resources. Recent developments in the field of AI open up entirely new possibilities for scalable implementation: Based on large language models, dialogue systems can be developed that interact linguistically with children in accordance with the principles of dialogic reading. The generation of individually tailored picture book stories and supplementary support is also, in principle, possible. In an interdisciplinary collaborative project, the necessary aspects of AI implementation and the conditions for success will be researched and empirically evaluated. To use the potential of AI-supported and scalable language support for daycare practice and improve children’s language development, close cooperation with education officials and daycare providers is planned.