AI Safety in China

AI Safety in China is a newsletter produced by Concordia AI (安远AI), a Beijing-based social enterprise focused on AI safety and governance. The newsletter was launched on August 8, 2023, with the first issue published on August 24, 2023, as part of Concordia AI's broader mission to guide the governance of AI for a long and flourishing future. The newsletter covers three primary topic areas: technical safety and alignment research conducted by Chinese scholars and research institutions, China's domestic governance and policy initiatives aimed at reducing AI risks, and China's engagement with and positions on international AI governance mechanisms. Each issue synthesizes recent publications, regulatory developments, and expert commentary from China's AI ecosystem. Concordia AI was founded in 2020 by Brian Tse (谢旻希), who previously served as Senior Advisor to the Partnership on AI and is a Policy Affiliate at the Centre for the Governance of AI. The organization is registered in Beijing as 北京谋远咨询有限公司 (Beijing Moyuan Consulting Co., Ltd.) and holds a Singapore office. In 2023, it was certified as a social enterprise by the Beijing Civil Affairs Bureau, committing at least 35% of after-tax profits to projects with a public purpose, making it one of the only AI safety-focused social enterprises in China. By mid-2025, Concordia AI had grown to 12 full-time staff and maintained a network of over 100 affiliated researchers from leading AI institutions globally. In addition to the newsletter, the organization publishes annual State of AI Safety in China reports (150+ pages), hosts major convenings including forums at the World AI Conference (WAIC) in Shanghai and in Singapore, advises policymakers and frontier AI labs, runs AI safety fellowship programs in China, and maintains a Frontier AI Risk Monitoring Platform tracking models from 15 leading developers worldwide. The newsletter had surpassed 1,400 Substack subscribers and the WeChat channel 4,600 subscribers as of mid-2025, with readership spanning government officials, AI labs, and AI safety researchers.