SELF-PACED · 6 WEEKS · AI RED TEAM

AI Security & Adversarial Machine Learning

Adversarial machine learning, prompt injection, LLM security testing, model poisoning, AI-powered attack automation, and defending AI/ML systems in production environments.

Prompt InjectionLLM SecurityAdversarial MLModel PoisoningAI Red TeamRAG Security

Enroll Now Get Free Guidance →

Weeks

Red Team Focus

LLM

Security

Lifetime

Access

Curriculum

Full Course Outline

8 modules · Practical labs throughout · Real-world case studies

AI/ML Security Landscape

Threat model for AI systems, attack surface overview, regulatory landscape

Prompt Injection Deep Dive

Direct injection, indirect injection via RAG, system prompt extraction, multi-turn attacks

LLM Jailbreaking

Jailbreak taxonomy, character-based attacks, many-shot, fine-tune bypasses

RAG System Vulnerabilities

Document poisoning, context manipulation, embedding attacks, retrieval manipulation

Adversarial ML Attacks

Evasion attacks, poisoning attacks, model inversion, membership inference

AI Red Teaming Methodology

Structured AI red team process, responsible disclosure, reporting for AI vulns

Defending AI Systems

Input validation, output filtering, monitoring, adversarial training, model hardening

AI Security Tools

PyRIT, Garak, custom testing frameworks, automated AI security scanning

Dr. Priya Sharma

GCIH · GCFA · PhD Computer Science (MIT)

Former Google Project Zero researcher. World-leading AI security expert. Published 14 CVEs across AI/ML systems. PhD from MIT. Author of widely-cited research on adversarial LLMs.

Enroll in This Course Get Free Career Guidance →

Who This Is For