Skip to content
NerdLevelTech
Intermediate

AI Security Fundamentals: Attacks, Defenses & Guardrails

Master AI security with OWASP LLM Top 10 vulnerabilities, prompt injection attacks, and production guardrails. Learn to defend LLM applications using NeMo Guardrails, LLaMA Guard, and security best practices.

65 min24 lessons6 modulesCybersecurity

What you'll learn

  • Identify and mitigate OWASP LLM Top 10 vulnerabilities
  • Detect and defend against prompt injection attacks
  • Implement NeMo Guardrails for LLM input/output filtering
  • Deploy LLaMA Guard for content safety classification
  • Design defense-in-depth strategies for AI applications

Course Content

  1. The AI Security Landscape

    4 lessons · 10 min
    • Why AI Security MattersNo signup3 min
    • Locked until you enrollOWASP LLM Top 10 Overview3 min
    • Locked until you enrollAttack Surface of LLM Applications2 min
    • Locked until you enrollSecurity Mindset for AI Developers2 min
    • Quiz: Module 1: The AI Security Landscape

  2. Prompt Injection Attacks

    4 lessons · 12 min
    • Locked until you enrollDirect Prompt Injection3 min
    • Locked until you enrollIndirect Prompt Injection3 min
    • Locked until you enrollAdvanced Injection Techniques3 min
    • Locked until you enrollJailbreaking vs Prompt Injection3 min
    • Quiz: Module 2: Prompt Injection Attacks

  3. Other Critical Vulnerabilities

    4 lessons · 10 min
    • Locked until you enrollSystem Prompt Leakage3 min
    • Locked until you enrollSensitive Data Exposure2 min
    • Locked until you enrollExcessive Agency & Permissions3 min
    • Locked until you enrollRAG & Embedding Vulnerabilities2 min
    • Quiz: Module 3: Other Critical Vulnerabilities

  4. Building Input/Output Guardrails

    4 lessons · 12 min
    • Locked until you enrollInput Validation Strategies3 min
    • Locked until you enrollOutput Sanitization3 min
    • Locked until you enrollNeMo Guardrails Setup3 min
    • Locked until you enrollLLaMA Guard & PromptGuard3 min
    • Quiz: Module 4: Building Input/Output Guardrails

  5. Production Security Patterns

    4 lessons · 12 min
    • Locked until you enrollDefense in Depth Architecture3 min
    • Locked until you enrollMonitoring & Logging3 min
    • Locked until you enrollRate Limiting & Abuse Prevention3 min
    • Locked until you enrollSecure Agent Design3 min
    • Quiz: Module 5: Production Security Patterns

  6. Security Testing & Next Steps

    4 lessons · 9 min
    • Locked until you enrollTesting Your Guardrails3 min
    • Locked until you enrollSecurity Checklists2 min
    • Locked until you enrollStaying Current2 min
    • Locked until you enrollNext Steps2 min
    • Quiz: Module 6: Security Testing & Next Steps