NTU AI Safety
Shaping the Future of
AI Safety
We are a student-led community at National Taiwan University exploring AI alignment, mechanistic interpretability, and AI governance.
Programs & Events
Join our reading groups, fellowships, and workshops on AI safety.
2026 Fall — Technical Paper Reading Group
Read selected AI safety papers on threat models, reward hacking, scheming, evaluations, mechanistic interpretability, and AI control.
Fundamental2026 Fall — Fundamental Reading Group
Explore AI alignment, safer training, evaluations, interpretability, and AI control through a semester of guided discussions.
Policy2026 Fall — Policy Reading Group
Discuss AI governance through state power, compute, Taiwan, China, institutions, and international cooperation.
Impact in the Making
From weekly reading groups to workshops, we've spent the past few years building something real: a community of people in Taiwan who take AI Safety seriously, think rigorously, and act. These are the moments that got us here.
Why AI Safety?
Advanced AI systems may soon surpass human capabilities across many domains. Ensuring they remain aligned with human values is one of the most important technical and governance challenges of our time.
NTUAIS brings together students, researchers, and practitioners to study these problems and build the next generation of AI safety talent in Asia.
Collaborated with
