The Singapore Consensus on Global AI Safety Research Priorities

TL;DR

This paper presents the Singapore Consensus on AI safety research priorities, organizing key challenges into development, assessment, and control to guide global research efforts in trustworthy AI.

Contribution

It synthesizes international AI safety research priorities into a structured framework based on a defence-in-depth model, guiding future research directions.

Findings

Organized AI safety challenges into three domains: Development, Assessment, and Control.

Provides a global consensus framework endorsed by 33 governments.

Highlights the importance of a layered approach to AI safety research.

Abstract

Rapidly improving AI capabilities and autonomy hold significant promise of transformation, but are also driving vigorous debate on how to ensure that AI is safe, i.e., trustworthy, reliable, and secure. Building a trusted ecosystem is therefore essential -- it helps people embrace AI with confidence and gives maximal space for innovation while avoiding backlash. The "2025 Singapore Conference on AI (SCAI): International Scientific Exchange on AI Safety" aimed to support research in this space by bringing together AI scientists across geographies to identify and synthesise research priorities in AI safety. This resulting report builds on the International AI Safety Report chaired by Yoshua Bengio and backed by 33 governments. By adopting a defence-in-depth model, this report organises AI safety research domains into three types: challenges with creating trustworthy AI systems (Development), challenges with evaluating their risks (Assessment), and challenges with monitoring and intervening after deployment (Control).