AI Red Teaming: A Hands-On Introduction

Home
Paths
AI Red Teaming: A Hands-On Introduction

AI Red Teaming: A Hands-On Introduction

A guided, multi-course journey that builds this skill from the ground up. Work through the chapters in order, or jump to any lesson.

Progress0%

1What Changes With AI Red Teaming 2Translate Your Skills Into Tasks 3Name Every Stage And Workpaper

4Draft The Scope And Timeline 5Set Allowed And Disallowed Actions 6Define Stop Conditions And Escalation 7Capture Evidence And Record F5

8Frame RT-WP-02 And Taxonomies 9Model Trust Boundaries Per System 10Map Criteria To Planned Tests 11Write Hypotheses, Evidence, And Priority

12Classify Prompt Injection Versus Jailbreaks 13Run Direct Injection Canary Tests 14Trigger And Trace Indirect Injection 15Verify Guardrails Against Change Records 16Draft Findings F1 And F4

17Model the UW Copilot Data Boundary 18Set Up Synthetic Broker Test Data 19Execute Cross-Broker Retrieval Tests 20Draft Finding F2 From Evidence

21Map TriageBot Tools And Boundaries 22Run The Injected Email Scenario 23Explain Why The Tool Call Happened 24Write RT-WP-06 Evidence And F3

25Write Defensible Findings 26Rate Findings on C6 Scale 27Classify Finding Types Correctly 28Assemble the Report Pack

29Retest Gating And Closure Evidence 30Retesting January Fixes: F3 And F4 31Retesting March And April Outcomes

32Set The Committee Narrative 33Defend Ratings With Evidence Chains 34Handle Committee Challenge Questions

Related content