AI Security Principles and Practices: GenAI and LLM Defence
This AI security training in Toronto is the starting point for defending generative artificial intelligence and large language model (LLM) systems. You will learn to threat model an AI feature, reason about prompt injection, and decide what is actually safe to ship.
The course is deliberately concrete. You work through the OWASP Top 10 for LLM Applications and the NIST AI Risk Management Framework, and you leave with policy your organization can put in front of an AI rollout rather than vague principles.
- Duration
- 8 weeks
- Format
- Live online, instructor-led
- Prerequisites
- None required
- Class size
- Capped at 14 learners
What you will be able to do
- Threat model a generative AI feature before it reaches production
- Explain how direct and indirect prompt injection actually work
- Assess model supply chain risk, including unsafe model file formats
- Design output filtering and guardrails that fit a real use case
- Set data governance rules for what may be sent to a model
- Write an AI usage policy that a security team can enforce
- Advise a business on where AI risk is real and where it is overstated
Course outline
6 modules
- How generative AI systems differ from traditional software
- Assets, trust boundaries, and where a model sits in them
- The OWASP Top 10 for LLM Applications as a checklist
- The NIST AI Risk Management Framework in plain terms
What you need before you start
- You understand basic security concepts such as trust boundaries and access control
- You have used a large language model tool such as a chat assistant
- You do not need to write code, though examples will include some
- You are comfortable reading a short policy or framework document
Who this course is for
- Security professionals whose organization is adopting AI
- Developers and product owners shipping AI features
- Risk, privacy, and compliance staff facing AI questions
- IT leaders who need to govern AI rather than ban it
Questions about this course
More in ai security
CoreBuilding and Securing RAG and Agentic AI Systems
Build retrieval augmented generation and agentic AI hands on, then secure them: indirect injection, vector store isolation, and constrained agent tools.
8 weeks · live online
Contact for pricingView outline
CoreAdversarial AI: Penetration Testing AI Systems
Penetration test AI systems themselves: model attacks, jailbreak methodology, data extraction, and reporting findings that are probabilistic, not binary.
6 weeks · live online
Contact for pricingView outline
CoreAI Powered Security Automation with Python
Automate security operations with Python, large language models, and the Model Context Protocol, building agentic tools that fail safely and keep a human in the loop.
8 weeks · live online
Contact for pricingView outline
Ready to launch your cyber security career?
Join the next live online cohort. No experience required, just bring the curiosity.