Q1easy
Why is Human-in-the-Loop (HITL) recommended for high-stakes AI applications?
Back to Artificial Intelligence & Machine Learning
10 Questions Total
AI Ethics, Security & Privacy MCQs
Topic-wise practice questions for Artificial Intelligence & Machine Learning. Test your knowledge, check solutions on demand, and pace your preparation.
Zero-leak answer engine • Rate-limited explanations
Practice in Timed ModeQ2medium
A researcher wants to train a model on sensitive patient data but is not allowed to move the data out of the hospital's secure server. Which approach allows training the model without moving the raw data?
Q3medium
An AI system for loan approval is found to consistently reject candidates from a specific zip code regardless of their financial data. This is an example of:
Q4easy
What is Explainable AI (XAI) primarily used for?
Q5easy
Which of these is a key Privacy concern in modern LLMs?
Q6medium
Which principle states that an AI system should be understandable by its creators and users?
Q7medium
A model developer runs an audit. Based on the confusion matrix below, which group is being unfairly targeted?
Q8medium
An attacker adds tiny, invisible pixels to an image of a cat, causing the AI to classify it as a "toaster." This is an example of what?
Q9medium
A malicious actor introduces thousands of fake, mislabeled photos into a training dataset to intentionally degrade the model's performance. This is:
Q10easy
What is Differential Privacy?
audit_matrix.log
Group A: 95% Accuracy
Group B: 60% Accuracy