What is a correct way to reduce bias in Artificial Intelligence?
Answer and explanation
Correct answer: Using diverse, representative, and balanced data
Training data should represent the relevant groups and situations fairly, with quality checks for missing or distorted information. Diverse and balanced data can reduce the chance that a model systematically favours or harms one group, so option B is correct. It does not remove all bias automatically, but it is an important safeguard.
Frequently asked questions
What is the correct answer to this question?
Using diverse, representative, and balanced data
Why is this the correct answer?
Training data should represent the relevant groups and situations fairly, with quality checks for missing or distorted information. Diverse and balanced data can reduce the chance that a model systematically favours or harms one group, so option B is correct. It does not remove all bias automatically, but it is an important safeguard.
Which subject and chapter does this question cover?
This is a Class 9 Artificial Intelligence question. Chapter: Reflection, Project Cycle, and Ethics. Topic: AI Ethics.
