AI features introduce risks that traditional security training does not cover. This AI security training teaches you how applications built on large language models can be attacked, how to test them and how to design them more safely.
Who should attend
- Security engineers and penetration testers
- Developers and architects building AI-powered features
- Product and engineering leads responsible for AI risk
- Risk and compliance teams that need to understand AI threats
- Penetration testers new to application testing can start with our Web Application Penetration Testing Training.
What you will learn
- How LLM-based applications are built: prompts, retrieval, tools and agents
- The OWASP Top 10 for LLM Applications
- Direct and indirect prompt injection
- Jailbreaks and guardrail bypass
- Sensitive data disclosure and system prompt leakage
- Insecure output handling and downstream impact
- Excessive agency: agents, tools and function calling risks
- RAG and vector database security
- Abuse, cost and resource consumption risks
- Threat modeling AI systems
- Testing approaches and reporting AI security findings
- Design patterns that limit the impact of successful attacks
Hands-on labs
Participants work with intentionally vulnerable AI applications in a lab, practicing both testing and defensive design. Exercises cover chatbots, retrieval-augmented generation (RAG) pipelines and agents with tool access.
What you receive
- Course materials and lab access
- An AI application security checklist
- A certificate of attendance
Frequently asked questions
Learn to build and test AI securely
Contact us for dates, pricing and corporate options.
Need 24/7 Protection From Cyber Attacks?