GENERATIVE ARTIFICIAL INTELLIGENCE AND PROBLEM-BASED LEARNING IN MATHEMATICS EDUCATION: A SYSTEMATIC LITERATURE REVIEW
TL;DR
Investigation of how AI has been used to promote critical thinking, argumentation, and complex problem-solving within the context of mathematics education or STEM fields concludes that the integration of AI into PBL requires pedagogical intentionality.
Abstract
The integration of Generative Artificial Intelligence (GenAI) into Problem-Based Learning (PBL) environments presents new challenges and opportunities for education. The objective of this research is to investigate how AI has been used to promote critical thinking, argumentation, and complex problem-solving within the context of mathematics education or STEM fields. To this end, a Systematic Literature Review (SLR) was conducted, guided by the PRISMA protocol, analyzing empirical studies published since 2020 and indexed in the Scopus, IEEE Xplore, Web of Science, and ERIC databases. The results indicate that AI transcends the function of a content repository, assuming pedagogical roles such as 'Socratic Coach,' 'Design Partner,' and 'Object of Critical Analysis.' The analysis highlights that the tool's effectiveness depends intrinsically on the quality of prompt engineering and human mediation. Furthermore, AI 'hallucinations' and mathematical inconsistencies do not necessarily need to be perceived as critical flaws; they can emerge as effective pedagogical catalysts for the refinement of logical validation by students. It is concluded that the integration of AI into PBL requires pedagogical intentionality, shifting from prohibitive models to approaches that foster digital literacy and academic integrity within a hybrid intelligence ecosystem.