Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification
Recent studies demonstrate that machine learning algorithms can discriminate based on classes like race and gender. This academic study presents an approach to evaluate bias present in automated facial analysis algorithms and datasets.
The study examines commercial gender classification algorithms and reveals significant disparities in accuracy based on gender and skin type.
Darker-skinned females experience the highest misclassification rates (up to 34.7%), while lighter-skinned males have much lower error rates (as low as 0.8%). The findings emphasize the importance of balanced datasets and intersectional evaluation to promote fairness and accountability in AI facial analysis systems.
Share this Resource:
