AI AI Engineer: AI System Design and Ethics 2 — Questions and Answers
Question 1: What is the 'right to explanation' under AI regulations like the EU AI Act?
- The right for AI engineers to document their models
- The right for individuals to receive a meaningful explanation of automated decisions that affect them (Correct answer)
- The obligation to explain AI research publicly
- The requirement to open-source AI training code
Correct answer: The right for individuals to receive a meaningful explanation of automated decisions that affect them
Regulations like the EU AI Act and GDPR grant individuals the right to understand why an automated system made a decision affecting them.
Question 2: What distinguishes a 'high-risk AI system' under the EU AI Act?
- AI systems with more than 1 billion parameters
- AI used in domains like hiring, credit, healthcare, or law enforcement where errors have significant societal impact (Correct answer)
- AI systems that run on public cloud infrastructure
- AI systems developed outside the EU
Correct answer: AI used in domains like hiring, credit, healthcare, or law enforcement where errors have significant societal impact
The EU AI Act classifies AI as high-risk based on use-case domain (e.g., employment, critical infrastructure, law enforcement) due to potential for significant harm.
Question 3: What is 'fairness through unawareness' and why is it insufficient?
- Ignoring all training data to avoid bias; insufficient because it produces random outputs
- Excluding protected attributes (e.g., race) from model inputs; insufficient because proxies still encode the information (Correct answer)
- Balancing classes in training data; insufficient because it doesn't address test-time bias
- Using equal sample sizes per group; insufficient because it ignores distribution differences
Correct answer: Excluding protected attributes (e.g., race) from model inputs; insufficient because proxies still encode the information
Excluding protected attributes doesn't prevent discrimination because correlated proxy variables (zip code, name) still allow the model to infer and act on them.
Question 4: What is 'data governance' in the context of AI system development?
- Governing GPU resource allocation during training
- A framework of policies and processes for ensuring data quality, privacy, security, and compliance throughout the AI lifecycle (Correct answer)
- Controlling who can deploy AI models
- Managing version control for training scripts
Correct answer: A framework of policies and processes for ensuring data quality, privacy, security, and compliance throughout the AI lifecycle
Data governance defines who owns data, how it's collected and used, retention policies, and ensures compliance with privacy regulations throughout the AI pipeline.
Question 5: What is an 'AI incident' as defined in responsible AI frameworks?
- A model failing to converge during training
- An event where an AI system causes or contributes to harm, near-miss, or unexpected negative consequences in deployment (Correct answer)
- A disagreement between AI engineers about model architecture
- A drop in model accuracy below a threshold
Correct answer: An event where an AI system causes or contributes to harm, near-miss, or unexpected negative consequences in deployment
An AI incident is any real-world event where an AI system causes harm or poses significant risk, tracked in repositories like the AI Incident Database.
Question 6: What does 'consent' mean in the context of using personal data to train AI models?
- Getting approval from the AI ethics board
- Obtaining informed agreement from individuals before using their personal data for model training (Correct answer)
- Signing an NDA with data providers
- Having executives approve the training dataset
Correct answer: Obtaining informed agreement from individuals before using their personal data for model training
Consent requires that individuals knowingly agree to how their personal data will be used, including for AI training, per regulations like GDPR and CCPA.
AI AI Engineer: AI System Design and Ethics 2
What is the 'right to explanation' under AI regulations like the EU AI Act?