Towards an international regulatory framework for AI safety: lessons from the IAEA's nuclear safety regulations

This study explores the necessity and direction of safety regulations for Artificial Intelligence (AI), drawing parallels from the International Atomic Energy Agency (IAEA)'s nuclear safety regulations. The rapid advancement and global proliferation of AI necessitate standardized safety norms to minimize discrepancies between national regulations and enhance consistency and effectiveness. International collaboration and stakeholder engagement are crucial for appropriate and continuously updated regulations. The study highlights the risks of improperly tuned subgoal setting mechanisms in AI's decision-making, advocating for regulatory approaches to guarantee safe and predictable AI operations. It acknowledges limitations of directly applying IAEA models to AI due to distinct characteristics and risks, calling for future research on a tailored framework.

Seokki Cha