UCLA Journal of International Law and Foreign Affairs

Thee pursuit of artificial general intelligence (AGI) poses existential risks to humanity. Global cooperation is needed to mitigate these risks. This Article proposes objectives and ideal features of a global governance regime focused on unaligned AGI and other high-risk artificial intelligence (AI). It utilizes a risk-based approach to suggest mechanisms and legal frameworks that address two objectives: safety and accountability. Along with noting ideal characteristics of a global AI safety regime, this Article evaluates precedents in existing, comparable international systems to explore possible components of and entities for global governance of high-risk AI.