Problem-based and project-based learning as promising frameworks for generative AI-supported education: Emerging evidence from a systematic review and three-level meta-analysis.

Abstract

As AI advances, specifying which competencies learners will need grows harder, making it urgent to identify stable approaches for cultivating competencies that remain valuable regardless of which AI tools emerge next. Problem-based learning (PBL) and project-based learning (PjBL) may provide one suitable learning context for this: by integrating generative AI (GenAI) into authentic problem solving and meaningful production, they may help cultivate the judgment to evaluate AI outputs and the capacity to coordinate AI use across complex workflows, capacities that may remain relevant to competent work in future social and professional life. As an early-stage synthesis, this study brings together emerging evidence to clarify key concepts in this nascent field. A three-level meta-analysis of 22 controlled studies (66 effect sizes, January 2023-April 2026) found significant effects across four learner-internal outcome domains (learning achievement and skill performance, higher-order thinking, affective and motivational outcomes, and collaboration and communication outcomes), yielding an overall effect of g = 0.819. Sensitivity analyses indicated small-study effects, suggesting that the unadjusted effect may be inflated; the PEESE specification yielded a reduced estimate of g = 0.378. Exploratory moderator analyses indicated that pooled effect estimates differed significantly by participant number and GenAI tool type; peer collaboration showed a marginal trend. Product and solution performance, a traditional focus of PBL/PjBL assessment, was modeled separately; its large pooled estimate (g = 1.958) may reflect AI-augmented task performance, but its wide prediction interval crossed zero, indicating substantial uncertainty across settings. Overall, the evidence is promising but preliminary.