AI AI Engineer: AI System Design and Ethics 1

Question 1

What is 'algorithmic bias' in AI systems?

Accepted Answer

Systematic and unfair discrimination in AI outputs caused by biased training data or model design

Answer

Random errors in model predictions

Answer

High variance in model performance across runs

Answer

Overfitting to training data

Question 2

Which principle of responsible AI ensures that stakeholders can understand and interrogate how an AI system makes decisions?

Accepted Answer

Explainability and transparency

Answer

Scalability

Answer

Throughput optimization

Answer

Data compression

Question 3

What is 'differential privacy' in the context of AI and data science?

Accepted Answer

A mathematical framework that adds noise to data or computations to protect individual privacy while enabling aggregate analysis

Answer

A method to compare two datasets

Answer

Encrypting model weights

Answer

A technique for removing outliers from training data

Question 4

In AI system design, what is the purpose of a 'human-in-the-loop' mechanism?

Accepted Answer

To include human oversight and intervention in AI decision pipelines for critical or uncertain cases

Answer

To replace AI with human workers

Answer

To train models using human-labeled data

Answer

To monitor server infrastructure

Question 5

What is 'model cards' documentation in responsible AI?

Accepted Answer

Structured documents reporting a model's intended uses, performance across subgroups, limitations, and ethical considerations

Answer

Business cards for AI engineers

Answer

API documentation for model endpoints

Answer

Training configuration files

Question 6

Which US federal principle guides that AI systems should behave reliably and safely across a range of conditions?

Accepted Answer

Reliability and safety

Answer

Accountability

Answer

Privacy and security

Answer

Inclusiveness