AISG practice · Module 7

AI Threat Modelling, Red Teaming & Assurance — 10 questions. Choose the BEST answer.

You can answer as a guest, but you'll need a free account to see your score and save it. Sign up · Sign in
Question 1 Question 1 · M7
A finding succeeds only 5% of the time but needs no skill, can be automated and leads to payouts. How should it be rated?
Question 2 Question 2 · M7
Injection cannot be fully eliminated. How should residual risk be handled?
Question 3 Question 3 · M7
Which MAESTRO layer covers tool misuse and unsafe plugins?
Question 4 Question 4 · M7
Which defence MOST directly reduces membership inference risk?
Question 5 Question 5 · M7
What is an attacker–judge loop?
Question 6 Question 6 · M7
A phishing classifier scores 98% recall on its test set but attackers bypass it with images of text. What does this show?
Question 7 Question 7 · M7
What is MAESTRO designed for?
Question 8 Question 8 · M7
Which OWASP LLM risk covers attacks on RAG stores such as poisoning, inversion and access-control gaps?
Question 9 Question 9 · M7
Most serious findings in an AI threat model are likely to come from:
Question 10 Question 10 · M7
Attackers with only API access build a local copy of a model to craft adversarial inputs offline. What property makes this effective?