A visual walkthrough of how security responsibility changes when AI moves from answering questions to taking actions.
Learn why agentic AI changes the security model — and where controls, proof, and human oversight fit. A short interactive intro teaches the controls first.
← → move between beats — so does clicking the left or right half of the stage.
Hover a station for a quick inspector; click to pin the full one. Esc closes it.
Drag the slider (bottom left) to explore the five scopes on any beat.
Notes toggles speaker notes · ? opens this card.
Top bar: current beat, act, and state. Bottom left: the trust-boundary instrument — who owns what. Bottom right: the story panel; it becomes the inspector when you touch a station.
Station tint: green provider-leaning · amber increasingly yours.
You buy AI
1 Public AI App · consumer
2 Enterprise AI App
You build AI
3 Build on a Pretrained Model
4 Fine-Tuned Model
5 Train Your Own Model
An interpretive walkthrough of AI for Security and Security for AI (AWS + SANS, November 2025). 40 beats · 20–30 minutes · 300-level.