MCQ Single Best Answer Not Set

QWhat is model poisoning?

ID: #24840 AI Governance & Risk Management 5 views

Question Info

#24840Q ID

Not SetDifficulty

AI Governance & Risk ManagementTopic

Start Quiz on This Topic

Your Answer

Choose the Best Option

Click any option to instantly check if you're correct.

A Password reuse ✔ ✖
B Injecting malicious data into training ✔ ✖
C API throttling ✔ ✖
D Encrypting model weights ✔ ✖

Correct Answer: Option B

Explanation

The correct answer is: Injecting malicious data into training

Explanation:

Model poisoning is an attack where an adversary deliberately inserts malicious or manipulated data into the training dataset. This can cause the AI model to behave incorrectly, produce biased results, or even include hidden backdoors. Security frameworks like those from NIST highlight this as a key risk in AI systems.

Why not the others?

❌ Password reuse → A security issue, but unrelated to AI training
❌ API throttling → Used to limit requests, not an attack on training data
❌ Encrypting model weights → A protection method, not an attack

Continue Practice

Previous Question Next Question

Share This Question

Challenge a friend or share with your study group.