AI-WELL: Rethinking Sustainable Youth Athlete Development through Machine Learning in the Nordic Model of Athlete Education

The AI-WELL project aims to optimize student-athletesʼ development within the Nordic model of athlete education and generate data-driven insights to strengthen youth wellbeing more broadly. Sport schools have become an increasingly common way to combine competitive sport and schooling from age 13 at both lower- and upper secondary levels across the Nordic countries, yet little is known about their long-term effects on development and performance outcomes. Building on a unique longitudinal dataset of more than 800 Finnish student-athletes, the project will follow the longitudinal development of student-athletes during their years in lower secondary sport schools covering approximately 1,200 student-athletes across five Nordic countries. Using innovative machine learning techniques and a mixed-methods design, AI-WELL is organized into five work packages: (WP1) longitudinal analyses of adolescent development from lower to upper secondary school, (WP2) cross-cultural comparisons of sport school systems to identify best practices in athlete education across the Nordic region, (WP3) examination of gendered norms, wellbeing, and social belonging in sport school environments, (WP4) development and piloting of a digital wellbeing monitoring system to identify risk patterns among student-athletes, and (WP5) participatory action research to co-create practical and policy recommendations for advancing a sustainable Nordic model of athlete education. By combining high-quality longitudinal data with predictive machine learning models, the project has the potential to advance theory and methodology across disciplines and establish new approaches for analyzing complex datasets in psychology and educational science. Through its strong foundation, AI-WELL will generate knowledge with the potential to enhance youth wellbeing and health across the Nordic region and within diverse educational systems.