AWS Certified AI PractitionerResponsible AIMedium
A company is deploying an AI system for automated hiring. To adhere to Responsible AI best practices and prevent unintended biases from impacting job candidates, which human intervention strategy should be implemented?
- AFully automate the hiring process to remove human subjective bias.
- BIntegrate Human-in-the-Loop (HITL) to review critical AI decisions.
- CDisable all logging and auditing features to protect applicant privacy.
- DTrain the AI exclusively on historical hiring data to maintain consistency.
Show answer & explanationAnswer & explanation
Correct answer: B. Integrate Human-in-the-Loop (HITL) to review critical AI decisions.
Integrating Human-in-the-Loop (HITL) allows human oversight and intervention at critical stages of the AI's decision-making process, such as reviewing hiring recommendations. This helps catch and correct potential biases or errors that the AI might introduce, ensuring fairness.
Why the other options are wrong
- A. Full automation removes human oversight, which is contrary to preventing unintended biases in sensitive areas like hiring.
- C. Disabling logging and auditing makes it impossible to track, analyze, and address any biases or issues that arise, directly violating Responsible AI principles.
- D. Training exclusively on historical data is a known source of perpetuating and amplifying existing human biases present in past hiring decisions.
Human-in-the-Loop (HITL)
Human-in-the-Loop (HITL) is an approach to AI development and deployment that involves human intervention and oversight in critical decision-making processes.
- Combines human intelligence with machine learning.
- Crucial for tasks requiring judgment, nuance, or high stakes.
- Helps improve AI accuracy, fairness, and safety.
Memory trick: HITL is like having a co-pilot for your AI, always ready to take the wheel.