Harmonizing Governance Frameworks For The Agentic Use Of Al In Education Across The G20

The emergence of Artificial Intelligence (AI) presents both opportunities and challenges for education systems and practices, and work within G20 partner countries (European Commission 2022; Accenture 2023). The integration ofAl in education affords interesting new opportunities and capabilities for educators to customize learning experiences and promote higher order learning. However, the unsystematic adoption of Generative Al (GenAl) tools can create additional challenges and inequities for educational systems unprepared to address this new and rapidly evolving technology. This raises cause for concern on several fronts. Firstly, GenAl content raises issues ofprivacy, accuracy and authenticity ofcontent that challenges the integrity oftraditional assessment processes (Moorhouse et al. 2023). Secondly, the arbitrary use of GenAl content can result in lower levels of cognitive engagement, impacting the development of competent, self-directed learners. Thirdly, GenAl created content may reinforce misconceptions and biases and reduce the plurality of represented perspectives. Disparities in access to technology have the potential to further deepen existing digital and developmental divides. Lastly, GenAl has a substantial carbon footprint, which is of particular concern in terms of responding to the sustainable development goals ofAgenda 2030 (UN 2015). Educators and policy makers in G20 partner countries are actively considering the wider implications ofAl for education, with various artificial intelligence guidelines emerging from international organizations such as UNESCO (Miao & Holmes 2023), European Parliament and the United Nations, as well as bodies at national levels such as the National Academic Integrity Network in Ireland (NAIN 2022). This policy brief recommends the harmonization of governance frameworks regulating Al integration in education across G20 partner countries.