We focus on strengthening the local research ecosystem by nurturing emerging talent and providing the tools and support needed for career development in AI safety. This includes facilitating knowledge acquisition, creating meaningful opportunities, and improving accessibility, with the aim of fostering long-term growth and advancement within the community.
Our work advances frontier AI safety as a sociotechnical challenge by integrating multiple disciplines, including technical AI safety, cognitive science, law, and sociology. Through this interdisciplinary approach, we contribute to the development of safe and trustworthy AI systems.
Given the global nature of AI risks, collaboration across borders and disciplines is essential. We actively engage with an international community of AI safety researchers, practitioners, and policymakers, contributing to an interdisciplinary network that spans academia, industry, and governance.
Can debate guide human and LLM supervisors toward the truth despite their prior beliefs? Across experiments with 507 human participants and several LLM judges, debate raises human accuracy from 41.3% to 57.6%, while weaker language models show no comparable gains, raising concerns about fully automated oversight pipelines.
We present the "Pre-registration and REporting Protocol for AI Evaluations" (PREP-Eval), a step-by-step guide for planning and conducting AI evaluations. Inspired by practices in software testing, data mining, and psychology, PREP-Eval incorporates pre-registration, and its application is demonstrated across six diverse use cases.
Across five experiments with human participants, this research provides evidence on how people perceive the causal contribution of AI in both misuse and misalignment scenarios, and how these judgments interact with the roles of users and developers. These findings can not only inform the design of liability frameworks for AI-caused harms but also help explain or anticipate future related decisions.
How the press has covered our research.
