Microsoft Azure AI Fundamentals (AI-900)Describe AI workloads and considerationsMedium
A research team is developing an AI model to predict the outbreak of rare diseases based on environmental factors and anonymized patient data. It is crucial that the model's predictions are consistently accurate and that erroneous predictions do not lead to harmful public health decisions. Which responsible AI principle is most directly concerned with the model's consistent performance and safety?
- AReliability and Safety
- BInclusiveness
- CFairness
- DPrivacy and Security
Show answer & explanationAnswer & explanation
Correct answer: A. Reliability and Safety
Reliability and Safety ensure that AI systems operate consistently, accurately, and without causing harm. In a public health context, consistent accuracy and preventing harmful decisions are direct applications of this principle.
Why the other options are wrong
- B. Inclusiveness focuses on accessibility and empowering diverse users, not the inherent accuracy and safety of the model's output.
- C. Fairness addresses bias and equitable treatment, not the operational consistency or safety of predictions.
- D. Privacy and Security protect data, which is important but not the primary concern for the model's predictive performance and safety.
Responsible AI: Reliability and Safety
Ensuring AI systems perform consistently, accurately, and safely, avoiding unintended harm.
- Focuses on robustness, accuracy, and error handling.
- Crucial in high-stakes applications like healthcare and autonomous systems.
- Involves rigorous testing and validation.
Memory trick: AI must be RELIABLE and SAFE, like a well-built machine.