Scaling Project-Based Learning with Artificial Intelligence-Powered Parametric Feedback—A Model for Nakhchivan State University

Project-based learning (PBL) is a vital pedagogical approach that cultivates critical thinking, collaboration, and problem-solving skills. However, its widespread adoption in higher education, particularly in contexts like Azerbaijani universities, is hindered by excessive faculty workload, difficulties in providing continuous assessment, and the absence of scalable digital scaffolding. This study proposes and validates an institution-based model for Nakhchivan State University that utilizes Artificial intelligence (AI) as a parametric feedback mechanism to streamline and scale PBL application. The AI is conceptualized as an adaptive system that maximizes the flow of information and tunes support levels, operating within a hybrid intelligence (HI) framework. Methodologically, the study introduces an AI-supported parametric feedback architecture that integrates learning analytics, rubric-based evaluation, and adaptive feedback loops within a HI framework and empirically validates its pedagogical effectiveness through an eight-week pilot implementation. A pilot study involving 304 students and 15 faculty members in the Faculties of Education and Engineering demonstrated empirical validity. Outcomes show that AI integration substantially reduced faculty workload (26– 28% reduction), resolving the logistical barrier to PBL scalability. Simultaneously, the system produced meaningful gains in student outcomes: project quality improved by 12 points on the standardized rubric, with category-specific gains of +13 in technical accuracy. This improvement was strongly correlated with student perceptions that the AI feedback was actionable and relevant. Furthermore, the model significantly enhanced collaboration and self-regulation. Collaboration analytics documented a 23.7% increase in mean individual contributions per student, mitigating social loafing and unequal participation. The continuous, adaptive feedback fostered Self-Regulated Learning, with over 80% of students reporting high confidence in managing projects and deadlines. Crucially, the ethical integration was affirmed by high trust levels; over 82% of students found the AI feedback unbiased and transparent. In conclusion, this AI-PBL model provides an empirically validated framework for the responsible and effective scaling of PBL in higher education, demonstrating that AI can both resolve institutional bottlenecks and significantly enhance core student competencies and collaborative equity, all while maintaining strong pedagogical alignment with constructivist learning theories. Cite this article as: R. Jafarli, G. F. Jafarli, A. Özperçin and E. Adıgüzel, "Scaling project-based learning with artificial intelligence–powered parametric feedback—A model for Nakhchivan State university," Electrica, 2026, 26, 0054, doi: 10.5152/electr.2026.26054.