Which concept is about ensuring AI systems are not biased toward a particular group due to faulty data?

• A. Bias in AI
• B. Fairness in AI
• C. Transparency in AI
• D. Privacy in AI

Answer: (A)

Bias in AI is about how models can inherit and amplify biases from the data they were trained on. If the training data are faulty, unrepresentative, or labeled in biased ways, the model’s predictions can systematically favor or disadvantage a particular group. The scenario in the question points to this exact issue: preventing the system from being biased toward a group due to faulty data.

Fairness in AI is related but broader—it focuses on ensuring equitable outcomes across groups, which may involve techniques to reduce bias, but it’s not the specific label for biases arising from data quality. Transparency in AI and Privacy in AI address different concerns: how decisions are made, and protecting personal data, respectively.