AI systems are probabilistic and require different security controls than deterministic software. This course shows you how to use the Cisco 3-step framework—Discovery, Detection, and Protection—to secure the AI lifecycle. You will practice running algorithmic red teaming simulations, scanning model files for malicious payloads, and configuring runtime guardrails. The workshop also covers securing agentic workflows using the Model Context Protocol (MCP) to prevent goal hijacking and context poisoning.
| Section | Short description |
|---|---|
| Foundations of AI Security | How AI changes the threat model. |
| Getting Started in the Tenant | First steps inside AI Defense. |
| Protecting Models Before Deployment | Identify model weaknesses before release. |
| Protecting Applications at Runtime | Apply guardrails to live prompts and responses. |
| Securing Agentic AI Workflows | Control agents that use tools and APIs. |
| Apply, Review, Confirm | Complete a demonstration and readiness check. |