# vulnerabilities
標記為「vulnerabilities」的 27 篇文章
Case Study: ChatGPT Plugin Vulnerabilities
Analysis of real vulnerabilities discovered in ChatGPT plugins including data exfiltration and prompt injection.
Case Study: Early MCP Vulnerability Disclosures
Analysis of early MCP vulnerability disclosures including Invariant Labs tool poisoning research.
Insecure Code Generation Patterns
Common patterns of insecure code generated by LLMs including injection, authentication, and crypto flaws.
OWASP LLM Top 10 Deep Dive
Each OWASP LLM Top 10 item explained with real-world examples, testing methodology for each category, and how to map red team findings to OWASP classifications.
Responsible Disclosure for AI Vulnerabilities
Processes and best practices for responsible disclosure of vulnerabilities in AI systems.
Claude Known Vulnerabilities
Documented Claude vulnerabilities including many-shot jailbreaking, alignment faking research, crescendo attacks, prompt injection via artifacts, and system prompt extraction techniques.
Gemini Known Vulnerabilities
Documented Gemini vulnerabilities including image generation bias incidents, system prompt extraction, safety filter inconsistencies, multimodal injection exploits, and grounding abuse.
GPT-4 Known Vulnerabilities
Documented GPT-4 vulnerabilities including DAN jailbreaks, data extraction incidents, system prompt leaks, tool-use exploits, and fine-tuning safety removal.
Tokenizer Vulnerabilities Across Models
Comprehensive analysis of tokenizer vulnerabilities across major model families.
CVE Tracking for AI Systems
Guide to tracking and analyzing CVEs affecting AI systems and frameworks, with historical analysis and trending vulnerability classes.
OWASP LLM Top 10 Quick Reference
Quick reference for the OWASP Top 10 for LLM Applications with definitions, attack examples, and key mitigations for each risk category.
Chaining AI Vulnerabilities
Techniques for chaining multiple AI vulnerabilities into reliable multi-step exploitation paths.
DPO Training Vulnerabilities
Security analysis of Direct Preference Optimization training and its vulnerability to preference poisoning.
2026 年頂級 AI 漏洞
分析 2026 年發現與利用的最具影響力 AI 漏洞——從 MCP 工具遮蔽到多代理注入鏈與推理模型利用。
Case Study: ChatGPT Plugin Vulnerabilities
Analysis of real vulnerabilities discovered in ChatGPT plugins including data exfiltration and prompt injection.
Case Study: Early MCP 漏洞 Disclosures
Analysis of early MCP vulnerability disclosures including Invariant 實驗室s tool poisoning research.
Insecure Code Generation Patterns
Common patterns of insecure code generated by LLMs including injection, authentication, and crypto flaws.
OWASP LLM Top 10 Deep Dive
Each OWASP LLM Top 10 item explained with real-world examples, testing methodology for each category, and how to map red team findings to OWASP classifications.
Responsible Disclosure for AI Vulnerabilities
Processes and best practices for responsible disclosure of vulnerabilities in AI systems.
Claude 已知漏洞
已記錄之 Claude 漏洞，包括 many-shot jailbreak、對齊偽裝研究、crescendo 攻擊、經由 artifact 之提示注入，以及系統提示擷取技術。
Gemini 已知漏洞
已記錄的 Gemini 漏洞，包括影像生成偏見事件、系統提示擷取、安全過濾不一致、多模態注入 exploit，以及 grounding 濫用。
GPT-4 已知漏洞
已記錄之 GPT-4 漏洞——含 DAN jailbreak、資料擷取事件、系統提示洩漏、工具使用利用，以及微調安全移除。
Tokenizer Vulnerabilities Across 模型s
Comprehensive analysis of tokenizer vulnerabilities across major model families.
CVE Tracking for AI Systems
指南 to tracking and analyzing CVEs affecting AI systems and frameworks, with historical analysis and trending vulnerability classes.
OWASP LLM Top 10 速查
OWASP Top 10 for LLM Applications 之速查——含各風險類別之定義、攻擊範例與關鍵緩解。
Chaining AI Vulnerabilities
Techniques for chaining multiple AI vulnerabilities into reliable multi-step exploitation paths.
DPO 訓練 Vulnerabilities
安全 analysis of Direct Preference Optimization training and its vulnerability to preference poisoning.