Chapter C · 8 pages · ~104 min
Prompt injection & AI security
A practical deep dive into Prompt injection: mechanism, controls, tradeoffs, failures, evaluation, and an end-to-end decision trace.
What you will be able to do
- Explain Prompt injection using an accurate input-to-output mental model
- Choose controls for Prompt injection and predict their quality, latency, cost, and risk effects
- Diagnose realistic Prompt injection failures and design proportionate safeguards
- Evaluate Prompt injection against a baseline with task-specific and operational evidence