AI-Powered Discourse in Mathematics Education in Support of SDG 4: A Systematic Review of Contemporary Research Literature

Abstract This systematic review critically examines contemporary research on the integration of artificial intelligence (AI) in supporting mathematical discourse, with a specific focus on its alignment with Sustainable Development Goal 4 (SDG 4); which advocates for inclusive and equitable quality education. Drawing on 25 peer-reviewed studies published between 2015 and 2025, the review synthesises empirical and theoretical insights into how AI technologies such as natural language processing tools, intelligent tutoring systems, conversational agents and discourse analytics dashboards -transform mathematics education by enhancing student engagement, reasoning and participation. The findings reveal that AI-powered discourse interventions support multiple SDG 4 targets including improving learning outcomes (Target 4.1), promoting equity and inclusion (Target 4.5), and strengthening teacher capacity through professional development (Target 4.c). However, the review also identifies persistent challenges related to technological limitations, ethical concerns surrounding data privacy and algorithmic bias, and the need for extensive teacher training. Geographically, research remains predominant in East and Southeast Asia, North America, and the Middle East, with notable underrepresentation from the Global South. The review concludes by offering strategic recommendations for inclusive AI design, teacher-centred implementation and policy infrastructure that collectively aim to harness AI’s potential for fostering equitable and dialogic mathematics learning environments worldwide.