๐๐We are officially wrapping up our second AI safety intro fellowship and the first year of Hanoi AI Safety Network
(HAISN)! ๐ฅณ๐ฅณ

Our Journey
- Our group started as a small paper reading group with 2 people.
- We expanded city-wide to welcome talented students from other top tech universities such as Hanoi University of Science and Technology, Vietnam National University, Hanoi, and University of Science and Technology of Hanoi - USTH.
- We hosted intensive Sunday afternoon sessions, 3-4 hours each.
- We organised 2 iterations of the popular BlueDot Impact’s AI Safety Fundamentals courses
- Members engaged in numerous discussions, paper deep-dives, and research projects

Key Achievements
- ๐ 21 graduates from intro fellowships
- ๐ฌ 4 research fellows at Apart Research conducting AI safety research
- ๐ 3 publications at (Association for the Advancement of Artificial Intelligence (AAAI) Workshop on Datasets and Evaluators for AI Safety, ICLR under review)
- ๐ 8 hackathon projects, including a first-place win at the AI Security Hackathon
Notable Concrete Actions by Members
- โ๏ธ 7 members created technical content on various AI safety topics
- ๐ 6 members upskilled with the ARENA curriculum
- ๐ฃ๏ธ 5 members secured Open Philanthropy grant in collaboration with EA Vietnam for AI Safety content translation and outreach
- ๐ 4 members attended EAGx Singapore, networking with AI safety experts
- ๐ญ 1 member received a full-ride scholarship for Waseda University and is helping to establish an AI Safety research lab
Member Testimonials
- “HAISN got me to seriously consider the prospect that AI risks werenโt just sci-fi, and then gave me the resources to actually work on mitigating those risks.”
- โI was introduced to interesting research areas like mechanistic and developmental interpretability, and I think safety research can really use better engineering practices. Along with AI safety, I was also introduced to rationality and effective altruism, which is pretty cool.โ
Future Plans
- The group is seeking new leadership as Hieu Minh Nguyen pursues a research fellowship
- We are exploring innovative programme formats, including a “blog post fellowship” focusing on learning through writing
- Funding available for continued operations
๐ค Interested in collaborating or leading the group? Please reach out, we’d love to hear from you!