Introduction to LLM Vulnerabilities

What you'll learn

• Identifying LLM security vulnerabilities and attack vectors
• Mitigating model replication and shadowing attacks
• Recognizing insecure output handling and prompt injection
• Preventing model theft and excessive agency issues
• Implementing strategies for secure plugin design
• Redacting sensitive information using APIs and regex
• Monitoring and updating dependencies for security
• Analyzing generative AI application types and architectures
• Comparing API-based, embedded, and multi-model applications