Artificial Intelligence AI Ethics and Bias

Question 1

What is 'algorithmic bias' in AI systems?

Accepted Answer

Systematic errors in AI outputs that unfairly favor or disadvantage certain groups

Answer

The tendency of models to overfit training data

Answer

High variance in model predictions

Answer

Incorrect hyperparameter settings

Question 2

What does 'fairness' in AI typically require?

Accepted Answer

Equitable treatment and outcomes for individuals across demographic groups

Answer

Equal model accuracy across all users regardless of group

Answer

Models that are free from any statistical patterns

Answer

Identical outputs for all inputs

Question 3

What is 'explainability' (or interpretability) in AI?

Accepted Answer

The degree to which humans can understand and trace how an AI makes decisions

Answer

The ability to run models on edge devices

Answer

A model's accuracy on held-out test data

Answer

The speed at which a model produces outputs

Question 4

What is 'data privacy' in the context of AI model training?

Accepted Answer

Ensuring that personal data used in training is handled, stored, and used in accordance with privacy laws

Answer

Encrypting model weights

Answer

Using only public datasets

Answer

Removing duplicate records from training data

Question 5

What is the 'right to explanation' in AI regulation?

Accepted Answer

The right of individuals to receive a meaningful explanation of automated decisions affecting them

Answer

A requirement for AI companies to publish source code

Answer

The right to request model retraining

Answer

A requirement for AI models to be open source

Question 6

What is 'model transparency' in responsible AI?

Accepted Answer

Openly documenting how a model works, what data it was trained on, and its limitations

Answer

Making model weights publicly downloadable

Answer

Ensuring the model has no parameters

Answer

Publishing all training code without exception