techCert › AI Security Professional

AI Security Professional Certification

Attack, defend and harden AI systems. Learn to identify and mitigate vulnerabilities in AI systems, preventing prompt injection, data poisoning, and model inversion.

8 lessons · 6 Hours · Advanced · exam: 15 questions in 25 minutes, pass mark 70%.

What you will learn

• Map the AI attack surface using the OWASP Top 10 for LLM Applications
• Execute and defend against direct and indirect prompt injection
• Design least-privilege controls for AI agents and tools
• Detect and mitigate data poisoning and backdoors
• Defend models against extraction, inversion and membership inference
• Secure AI APIs, handle model output safely, and respond to incidents

Syllabus

Foundations

1. The AI threat landscape (30 min)

Prompt Injection

1. Understanding prompt injection (40 min)
2. Indirect injection & excessive agency (35 min)

Attacks on Models & Data

1. Defending against data poisoning (35 min)
2. Model extraction, inversion & membership inference (35 min)

Securing AI Applications

1. Securing AI APIs (35 min)
2. Output handling & sensitive data (30 min)
3. Monitoring, red teaming & incident response (30 min)