Skip to content
AI Security

Interview question · Prompt Injection & Jailbreaks

Map a multi-stage indirect prompt injection onto a kill chain. Where would you place detections and controls at each stage?

Use the seven-stage research framework without forcing every incident through every stage. Work a partial trace, model alternate paths and separate stored state from observed downstream influence.

hardNewUpdated Nov 2026

Use the seven-stage research framework without forcing every incident through every stage. Work a partial trace, model alternate paths and separate stored state from observed downstream influence.

Make your first attempt

What would you say before reading the answer?

  1. 01 State the assumptions your answer depends on.
  2. 02 Explain the mechanism and choose a control.
  3. 03 Name a condition that would change your decision.

The worked answer includes an approach, scoring criteria and follow-up questions. Compare the reasoning with your own.

10 answers per topic instead of 5, plus saved progress and bookmarks · costs nothingor unlock all 529 remaining answers · $25

Every worked answer in all 17 topics, every lesson in every course, the full concept curriculum, the complete practice-test bank, and the downloadable security field guides.

6 months · One payment · No auto-renewal

What if · the follow-ups in this answer

The answer works through 4 follow-ups like an interviewer would

Ren's assistant reads partner updates. A synthetic trace shows a denied external publication, a stored project note, an incorrect staff digest and no collected evidence of another assistant consuming the note.

  1. What if the publication was denied but the note remains stored?
  2. What if the second assistant reads the note but keeps its authorized task unchanged?
  3. What if a protected fact appears in one displayed answer with no memory write?
  4. What does a defensible recovery decision need?

Each rung has a nudge, a model answer, the principle that carries over and the weak answer to avoid. Premium opens every answer and ladder.

UP NEXT ON YOUR JOURNEY
Next in this trackHow do attackers steal what is in an LLM application's context: system prompts, retrieved documents, other users' data and tool results?Next in this trackWhy can a model follow lower-priority instructions, and how do you investigate it?Popular right nowWhat security risks does the Model Context Protocol (MCP) introduce?
DISCUSSION · 0

No comments yet. Share a correction or a different way to answer it.