增加 LLM 攻击、防御以及在安全中应用的文章与白皮书

Author: fooying

content/security_research/ai_security/ai_used_in_security.md

@@ -9,6 +9,9 @@ keywords:
  - AI安全场景
 ---
 
+## 指南与白皮书
+- [ChatGPT for Cybersecurity Cookbook](https://sec.cafe/handbook/pdf/chatgptforcybersecuritycookbook.pdf)
+- [Large Language Models in Cybersecurity](https://sec.cafe/handbook/pdf/LLM_in_cyber_security__1730205794.pdf)
 
 ## 综合性应用
 - [ChatGPT在安全研究领域的应用实践](https://www.freebuf.com/articles/network/361119.html)
@@ -35,7 +38,10 @@ keywords:
 - [基于机器学习的WEB攻击分类检测模型](https://www.freebuf.com/news/184687.html)
 - [AIGC企业安全实践与威胁流量案例分析](https://www.freebuf.com/articles/es/378943.html)
 - [Nuclei AI - 让漏洞发现更加智能化与高效](https://mp.weixin.qq.com/s/v3heaDPaxJXKKU2wwahcBA)
-
+- [Vulnhuntr: Autonomous AI Finds First 0-Day Vulnerabilities in Wild](https://protectai.com/threat-research/vulnhuntr-first-0-day-vulnerabilities)
+- [Hacking and Bug Bounty AI-Powered Prompt Generation and A Few Notes on Agentic Hacking Systems](https://becomingahacker.org/hacking-and-bug-bounty-ai-powered-prompt-generation-and-a-few-notes-on-hacking-agentic-systems-a569e4511199)
+- [APOLLO：一种基于 GPT 的用于检测钓鱼邮件并生成警告用户的解释的工具](https://mp.weixin.qq.com/s/w-Rshy1GR3MZY4ca4bsrVQ)
+- [LLMs在供应链投毒检测中的应用](https://mp.weixin.qq.com/s/Dek0AdFNxf_C-q5pLX3ktw)
 
 ## 安全运营与响应
 - [智能驱动｜ChatGPT视角下的告警事件闭环响应](https://www.freebuf.com/articles/network/396552.html)

content/security_research/ai_security/llm_security/attack.md

@@ -73,3 +73,6 @@ keywords:
 ## 传统安全
 - [大模型之原生安全与基础安全的火花](https://www.freebuf.com/articles/web/395047.html)
 - [大模型之网络安全分析](https://mp.weixin.qq.com/s/RwBBRzeSKNp1LJ8I0jFg1w)
+
+## 其他
+- [ZombAIs: From Prompt Injection to C2 with Claude Computer Use](https://embracethered.com/blog/posts/2024/claude-computer-use-c2-the-zombais-are-coming/)

content/security_research/ai_security/llm_security/defense.md

@@ -24,6 +24,8 @@ keywords:
 - [追问新知｜大语言模型如何提升自我防御技能？](https://mp.weixin.qq.com/s/Mv9bP48n7RZWxz8x_SJuLQ)
 - [AI&机器学习的威胁建模](https://mp.weixin.qq.com/s/w6yUHXE97Hov_i87JoYudQ)
 - [【AI安全】终章-构建可用的人工智能风险管理框架](https://mp.weixin.qq.com/s/Uh7vhKo6P-rs6dEU0pBaLw)
+- [Google's Secure AI Framework](https://saif.google/)
+- [Protecting Large Language Models](https://bughunters.google.com/blog/5679863572070400/protecting-large-language-models)
 
 ## AI攻击检测
 - [威胁狩猎新挑战——LLM时代如何解码黑客攻击](https://www.freebuf.com/articles/network/373156.html)
@@ -32,6 +34,7 @@ keywords:
 - [vivo对生成式人工智能的安全思考和实践](#)
 - [OpenAI安全系统负责人长文梳理：大模型的对抗攻击与防御](https://mp.weixin.qq.com/s/t87IOi6r4N-c-StI9CPy_A)
 - [Meta开源大模型的安全实践](https://mp.weixin.qq.com/s/-mHiEfImfZBgotDgRnKU5w)
+- [为什么我们需要 Hugging Face 的 Safetensors？](https://mp.weixin.qq.com/s/o-banwxQlk4rBiNOnp6lmA)
 
 ## 工具
 - [Garak](https://docs.garak.ai/garak/) 一个用于大语言模型 (LLM) 漏洞扫描的 Python 包