The drive for international cooperation in regulating frontier artificial intelligence (AI) models has gained momentum, with Singapore becoming the latest nation to join the European Union and 19 other countries in advocating for heightened oversight and safety measures. This coalition aims to mitigate the risks associated with advancing AI technologies, which, if left unchecked, could pose significant safety and security threats on a global scale.
Singapore’s Prime Minister Lawrence Wong emphasized the necessity for collaborative international efforts to harness the benefits of AI while managing its risks. He highlighted the cross-border nature of AI risks, underscoring the importance of maintaining human control and oversight over these powerful systems.
The joint declaration, which was adopted during the United Nations General Assembly in New York, acknowledges the transformative potential of AI in enhancing lives, propelling scientific research, and boosting economic development. Nonetheless, it warns of the dangers posed by rapidly evolving frontier AI models that could lead to serious repercussions if not properly managed.
Signatories of the declaration are urging UN member states to build upon current international frameworks and consider establishing a specialized international institution focused on frontier AI. Such an entity could play a crucial role in setting standards, supporting verification processes, and coordinating responses as AI systems reach new capability levels.
Furthermore, the declaration calls on governments to work together to develop common AI standards, enhance transparency, and establish shared methods for reporting significant AI safety incidents. It also emphasizes the need for improved access to scientific expertise and trusted evaluation capabilities across different regions.
AI companies are being encouraged to reinforce their safety protocols by thoroughly testing systems before deployment and employing independent evaluations to identify potential risks. Notably, Singapore is the sole Southeast Asian country to endorse this declaration, while major AI powers such as the United States and China have yet to join the initiative. The declaration remains open for additional countries to endorse the call for stronger AI safeguards.