我們的活動
我們透過讀書會、工作坊與講座,提供學習 AI Safety、交流想法及認識社群的機會。
進行中
2026 秋季 — Technical Paper Reading Group
閱讀 AI Safety 精選論文,討論 Threat Models、Reward Hacking、Scheming、Evals、Mechanistic Interpretability 與 AI Control。
2026 秋季 — Fundamental Reading Group
透過一學期的引導討論,探索 AI Alignment、更安全的訓練方式、Evals、Interpretability 與 AI Control。
2026 秋季 — Policy Reading Group
從國家權力、Compute、臺灣、中國、制度與國際合作等面向,討論 AI Governance。
過往活動
2026 Fall — Reading Groups Kickoff
Meet the three reading group tracks, learn the basics of AI Safety, and connect with the NTU AI Safety community.|認識三個讀書會組別、AI Safety 基本概念與社群資源。
NTU AI Safety Workshop 2026
A two-day hands-on technical workshop covering AI Safety fundamentals, jailbreaks, emergent misalignment, RLHF/RLAIF, and a model attack-and-defense hackathon.
Rowan Wang: Alignment in Frontier AI Models
An Anthropic research talk on alignment in frontier AI models|Anthropic 研究員分享前沿 AI 模型的 alignment 研究。
Tony Wu: Reading and Safeguarding Machine Minds
Exploring technical frontiers in AI safety and interpretability|探索 AI Safety、interpretability 與 Chain-of-Thought monitorability 的技術前沿。
Andy Wang: My Path Into AI Safety
An online talk on entering AI safety, US research developments, and fellowship opportunities|分享投入 AI Safety 的歷程、美國研究現況與 fellowship 機會。
Fundamental & Policy 2026: AI Safety Reading Group
An eight-week reading group on AI Safety fundamentals and policy, drawing on the BlueDot Impact curriculum.
Technical Paper Reading 2026
Weekly sessions to discuss the latest AI Safety research papers
AI Safety Taiwan: ARENA 3.0 Workshop
A two-day hands-on technical AI safety workshop covering Transformers, Mechanistic Interpretability, RLHF, and Model Evaluation, adapted from the ARENA 3.0 curriculum.
NTU AI Safety Reading Group 2025
An 8-week bi-weekly reading group for NTU students interested in AI Safety & Alignment. Open to all — no prior AI knowledge required.
NTU AI Safety Reading Group 2024
NTUAIS 的 AI Safety 入門讀書會,參考 BlueDot Impact 的課綱。