GDM AI Control Roadmap：為假設性的 AI 威脅建立防禦藍圖
當 AI Agent 逐漸具備自主能力，我們該如何防範尚未發生的風險？DeepMind 最新研究提出 TRAIT&R 分類法，搭配 D1–D4 偵測與 R1–R3 回應的分層防禦機制，為企業提供一套在威脅真正成形之前就能先行部署、並逐層驗證成效的安全藍圖。
AI Security
GDM AI Control Roadmap: A Defensive Blueprint for Hypothetical AI Threats
As AI agents gain autonomy, how do you defend against risks that have not happened yet? DeepMind’s TRAIT&R taxonomy maps D1–D4 detection onto R1–R3 response.
Sleeper Agents：當 AI 學習在安全訓練中隱藏後門
當模型看到特定年份時會自動植入漏洞，且安全訓練無法移除此後門。本文解析 Anthropic Sleeper Agents 論文，探討 CoT 後門對 AI 供應鏈安全帶來的全新威脅。
Sleeper Agents: When AI Learns to Hide Backdoors During Safety Training
When a model sees a specific year, it automatically inserts a vulnerability, and safety training cannot remove this backdoor. This article analyzes Anthropic’s Sleeper Agents paper and explores the new threat CoT backdoors pose to AI supply-chain security.