Position:Communities of Practice can be used to Address Challenges in Regulation and Governance of Generative AI in South East Asian Countries

Abstract

The proliferation of open-source, open-weight, and quantized large language models (LLMs) presents a transformative opportunity to advance health equity in Lower and Middle-Income Countries (LMICs), particularly within the diverse landscape of South East Asia (SEA). However, this potential is shadowed by the risk of exacerbating health disparities if these models, predominantly trained on data from Upper-Income Countries (UICs), are deployed without careful consideration of regional demographics, health needs, and cultural contexts. While Western nations are advancing regulatory frameworks for generative AI, SEA countries are at a more nascent stage. This paper summarizes key challenges in AI regulation and governance identified at a recent SEA leadership summit. We argue that a "one-size-fits-all" approach is untenable for this heterogeneous region. Instead, we propose the establishment of a regional Community of Practice (CoP). This CoP would serve as a collaborative platform for sharing knowledge, co-creating standards, and building trust among stakeholders. It would facilitate regional regulatory sandboxes and work towards harmonized approval processes, ensuring that generative AI is developed and deployed safely, effectively, and equitably. We conclude with a call to action for AI researchers, developers, administrators, and healthcare professionals to contribute to this vital initiative.

Ähnliche Arbeiten