Open Community for AI Safety China
Join community
EventsResourcesAbout

About Us

Open Community for AI Safety China (OCASC) was founded by researchers and enthusiasts passionate about AI safety, dedicated to building a high-quality AI safety knowledge ecosystem in China.

Our Story

Open Community for AI Safety China (OCASC) began as a small study group, founded by a group of researchers and enthusiasts passionate about AI safety. We started by periodically reading and discussing frontier international AI safety research. As more like-minded people joined, the community grew to become the largest AI safety open community in the Chinese-speaking internet.

Our mission is to support an open, high-quality AI safety ecosystem in China by improving access to frontier AI safety research, fostering technical discussion, and building connections with the global AI safety network.

Today, the community is centered around WeChat groups and active across multiple platforms including Bilibili, Zhihu, LinkedIn, and GitHub, reaching over 10,000 followers, with 600+ community members and 20+ core contributors.

Milestones

July 2026Became an IASEAI AffiliateThe community became an affiliate of the International Association for Safe & Ethical AI (IASEAI), strengthening its connection to the global AI safety and ethics network.
July 2026BlueDot Course — China EditionBringing the internationally recognized BlueDot AI safety curriculum to China, with lectures delivered by senior AI safety experts.
April 2026Building an Open CommunityThe community was officially renamed to Open Community for AI Safety China. Bi-weekly seminars were established with invited scholars, while presence expanded across Bilibili, Zhihu and other platforms, surpassing 200 members.
March 2026AI Control HackathonIn collaboration with Apart Research, the community hosted China's first AI Control Hackathon in Shanghai. About 10 teams participated — 6.5% of global submissions — competing across Build, Break, and Improve tracks.
April 2025Starting as a Study GroupA group of scholars and researchers interested in AI safety came together to found the AI Safety Study Group, beginning periodic reading sessions and discussions on frontier research.