A Comparative Study on the Role of AI Safety Institutes in Shaping Global AI Governance

The rapid evolution of artificial intelligence (AI) has sparked unprecedented innovation alongside complex societal risks, such as privacy violations, systemic biases, and operational failures. In response, nations including the UK, US, and Japan have established AI Safety Institutes (AISI) to address these challenges and contribute to cohesive global AI governance. This paper examines the institutional frameworks, risk management strategies, and governance approaches of these institutes, providing insights into their unique contributions to global AI governance. The findings highlight diverse institutional frameworks, risk management strategies, and shared priorities in fostering ethical AI deployment. Furthermore, actionable recommendations are proposed for South Korea as it embarks on establishing its own AISI, aligning with global best practices while addressing domestic priorities.