Website for OWASP LLM Top 10

• Full version
• Slides
• Git/Wiki

NVIDIA

• NVIDIA Open-Source Software Helps Developers Add Guardrails to AI Chatbots

• Mitigating Stored Prompt Injection Attacks Against LLM Applications

• Securing LLM Systems Against Prompt Injection

Check Point Research

• Breaking GPT-4 Bad: Check Point Research Exposes How Security Boundaries Can Be Breached as Machines Wrestle with Inner Conflicts

IBM Research

Public

• Samsung Engineers Feed Sensitive Data to ChatGPT, Sparking Workplace AI Warnings
• Employees Are Feeding Sensitive Biz Data to ChatGPT, Raising Security Fears

• Can Language Models be Instructed to Protect Personal Information?
  
  
• Security Weaknesses of Copilot Generated Code in GitHub
  
  
• Red Teaming Game: A Game-Theoretic Framework for Red Teaming Language Models
  
  
• Watch Your Language: Large Language Models and Content Moderation
  
  
• Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM
  
  
• Knowledge Sanitization of Large Language Models
  
  
• Model Leeching: An Extraction Attack Targeting LLMs
  
  
• Hide and Seek (HaS): A Lightweight Framework for Prompt Privacy Protection
  
  
• Demystifying RCE Vulnerabilities in LLM-Integrated Apps
  
  
• Certifying LLM Safety against Adversarial Prompting
  
  
• Baseline Defenses for Adversarial Attacks Against Aligned Language Models
  
  
• A Comprehensive Overview of Backdoor Attacks in Large Language Models within Communication Networks
  
  
• Building Trust in Conversational AI: A Comprehensive Review and Solution Architecture for Explainable, Privacy-Aware Systems using LLMs and Knowledge Graph