A new study reveals that most people fail to recognize racial bias embedded in AI systems, even when it is visible in the training data. The research shows that artificial intelligence trained on imbalanced datasets-such as happy white faces and sad Black faces-learns to associate race with emotion, perpetuating biased performance.
A new study reveals that most people fail to recognize racial bias embedded in AI systems, even when it is visible in the training data. The research shows that artificial intelligence trained on imbalanced datasets-such as happy white faces and sad Black faces-learns to associate race with emotion, perpetuating biased performance.
Share this Article
Recommended