Skip to content
AI Security

Interview question · LLM Application Security

Your product team is adding an AI assistant to an existing SaaS product. What changes in the threat model?

The assistant becomes a new client of every API the product has, with a confused-deputy problem built in. The scored answer finds the new trust boundaries and keeps the old access model intact.

mediumNewPreparation tags: Salesforce, Microsoft, GoogleUpdated Nov 2026

The assistant becomes a new client of every API the product has, with a confused-deputy problem built in. The scored answer finds the new trust boundaries and keeps the old access model intact.

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 6 follow-ups like an interviewer would

A CRM vendor adds an AI summary of each customer account, built from notes, emails and tickets, visible to sales staff.

  1. What new risks does the feature add?
  2. What if the summary uses records the viewer has no permission to see?
  3. What if a customer's email contains instructions to the summariser?
  4. What about sending customer data to a model provider?
  5. What if enterprise customers ask how the feature is secured?
  6. What would your security review cover?

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 trackWrite a sanitiser that keeps links and images in streamed model Markdown inert until they are validated.Next in this trackCan a vector search API leak confidential information even when it never returns the documents themselves?Next in this trackCosine similarity decides what a RAG system retrieves. How would you make retrieval prefer trustworthy sources instead of just similar ones?
DISCUSSION · 0

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