Mitre
PulseAugur coverage of Mitre — every cluster mentioning Mitre across labs, papers, and developer communities, ranked by signal.
1 天有情绪数据
-
Microsoft Security Copilot使用AI代理进行自主威胁检测
Microsoft开发了一个动态威胁检测代理（DTDA），并将其集成到其Security Copilot中，旨在自主调查安全事件并生成新颖的警报。该代理利用统一的活动时间线、版本化的LLM提示合同以及规划器-执行器循环来发现隐藏的威胁。在评估中，使用GPT-5.4时，DTDA实现了80.1%的精确率，并将F1分数比基线方法提高了多达0.26个点，证明了其大规模识别被遗漏恶意活动的能力。
-
CyberSecQwen-4B: Small, specialized model offers local defense for cybersecurity
A new, specialized language model named CyberSecQwen-4B has been developed for defensive cybersecurity tasks. This model is designed to be small, runnable locally, and handle sensitive data without needing external APIs…
-
AI工具发现大量真实世界安全漏洞
四月份，FreeBSD 发布了八份安全公告，其中六个漏洞是由人工智能识别的。这凸显了人工智能系统发现真实世界安全缺陷的日益增长的趋势，尽管它们也产生了大量错误输出。其中一个特定的漏洞 CVE-2026-4747，与通用弱点枚举 (CWE) 相关，并可能涉及 Claude 等 AI 模型。
-
MITRE warns of increased cyber risks with AI in medical devices
MITRE has identified increasing cybersecurity threats associated with the integration of AI, cloud computing, and post-quantum technologies into medical devices. These advancements, while offering potential benefits, in…