2.2.2. Applying Principles to Real Scenarios

Understanding the principles requires recognizing which applies in specific situations. The exam tests your ability to match scenarios to principles—not definitions, but application.

Decision framework for scenario questions:

1. Is someone being treated differently based on who they are? → Fairness
2. Do users understand how the AI works or that it's AI? → Transparency
3. Is someone responsible for the AI's decisions? → Accountability
4. Is data protected appropriately? → Privacy & Security
5. Does the AI work safely and consistently? → Reliability & Safety
6. Does the AI work for everyone, including people with disabilities? → Inclusiveness

Fairness scenarios:

• Loan approval systems must not discriminate based on gender or ethnicity
• Hiring algorithms must treat candidates equitably regardless of race
• Insurance pricing shouldn't vary by protected characteristics

⚠️ Exam Trap: Questions about loan or hiring AI often test fairness. The correct factors to consider are gender, ethnicity, and race—NOT salary, credit history, or payment history (which are legitimate business factors).

Transparency scenarios:

• Users must understand AI system limitations
• Users should know when they're interacting with AI
• AI decisions should be explainable

Accountability scenarios:

• AI systems must meet legal and ethical standards
• Someone must be responsible for AI system behavior
• Mechanisms exist for addressing AI errors

Privacy & Security scenarios:

• Healthcare AI must protect patient data
• Financial AI must secure sensitive information
• Data collection should be minimized to what's necessary

Reliability & Safety scenarios:

• Autonomous vehicles must operate safely
• Medical AI must perform consistently
• AI shouldn't cause physical or emotional harm

Inclusiveness scenarios:

• AI must work for users with disabilities
• Voice interfaces should understand diverse accents
• Products should be accessible to all users