Guiding the governance of AI for a long and flourishing future

We tackle high-impact and underexplored AI safety and governance challenges. We publish reports and briefs, fill knowledge gaps, and enhance the field’s collective knowledge.

We host and engage in global AI safety and governance conferences, through which we facilitate the sharing of best practices in AI safety and support international coordination.

We advise leading AI companies and labs on risk management practices. We provide state-of-the-art guidance on identifying, evaluating, and mitigating risks while fostering robust governance for general-purpose AI models.

We contribute to China’s national AI standards and engage in a wide range of national and global AI policy consultations.

We evaluate and monitor frontier AI risks across key domains including cyber offense, biological risks, chemical risks, harmful manipulation, and loss of control. Our platform tracks model capabilities and safety performance, providing transparent and up-to-date risk assessments.