How to be convincing when talking to people about existential threat from AI
"AI Safety" 相关文章
Hardening against AI takeover is difficult, but we should try
OpenAI’s new safety tools are designed to make AI models harder to jailbreak. Instead, they may give users a false sense of security.
AI Safety at the Frontier: Paper Highlights of October 2025
A Guide To Being Persuasive About AI Dangers
Why Safety Constraints in LLMs Are Easily Breakable? Knowledge as a Network of Gated Circuits
Sable and Able: A Tale of Two ASIs
Legible vs. Illegible AI Safety Problems
GDM: Consistency Training Helps Limit Sycophancy and Jailbreaks in Gemini 2.5 Flash
AI Safety Camp 11
Open-weight training practices and implications for CoT monitorability
Cohere's chief AI officer says AI agents come with a big security risk
研究表明 AI 承压能力差：为了一口电，竟愿突破安全底线
Research Reflections
How Powerful AIs Get Cheap
Red Heart
Leaving Open Philanthropy, going to Anthropic
【ICML25】使用信息瓶颈理论为点云模型进行错误归因，为安全问题构建可解释工具
硅谷甄嬛传爆更，马斯克转发！Ilya动手那一夜，谁捅了奥特曼一刀？
微软AI主管苏莱曼：只有生物体才能拥有意识
