Singapore has proposed a pilot initiative aimed at improving global artificial intelligence safety by facilitating the reporting and sharing of AI-related incidents among countries. Presented by Prime Minister Lawrence Wong during an online meeting of global leaders on October 28, the initiative seeks to prevent serious AI failures from recurring by encouraging a culture of transparency and collective learning.
The proposal, which was discussed during a meeting co-hosted by Finnish President Alexander Stubb and Norwegian Prime Minister Jonas Gahr Støre, emphasizes a practical and flexible approach. Wong highlighted the importance of sharing lessons from serious AI incidents across borders, drawing parallels to the international aviation industry’s reporting system that enhances safety through shared knowledge of accidents and near misses.
Under the proposed framework, participating nations would start by agreeing on common categories for AI incidents and identifying the necessary information to be shared. This would involve exploring secure methods for collecting and exchanging reports, with the goal of building confidence and addressing practical challenges before establishing a broader international reporting system.
The initiative comes in response to growing concerns about incidents involving advanced AI systems, such as unexpected behaviors during testing and unintended interactions with external software. These incidents have raised calls for stronger oversight and international cooperation in managing AI risks. During the meeting, leaders from 31 countries discussed enhancing risk management, building confidence in AI safety and security, and developing common governance approaches.
Wong also advocated for smaller groups of interested countries from diverse regions to develop practical tools and frameworks, leveraging existing international efforts like the Singapore Consensus and the work of the United Nations-backed Independent International Scientific Panel on AI. This approach aims to support the responsible development of advanced AI systems while ensuring that the benefits of the technology remain accessible.
As countries continue to explore measures to improve AI safety and security, Singapore’s proposal offers a potential pathway to enhance international collaboration and prevent the recurrence of serious AI incidents.