Exploring AI Bias Through Case Studies

Lubna·

AI Bias

Preview

At a Glance

Grade Level9th Grade

SubjectAI Bias

Original Prompt

Assigned Examples for Your Group: Facial recognition trained with poor data: Fails to recognize authorized users and grants access to unauthorized individuals. Medical image diagnosis system trained only on images of male patients: Fails to detect conditions in female patients, leading to incorrect or missed diagnoses. Activity Instructions: As a group, research (using the internet if available, or discuss your ideas) real-world cases similar to these errors (e.g., issues with faci...

This learning resource focuses on real-world examples of AI bias, specifically through case studies involving facial recognition technology and medical diagnosis. The document introduces students to the implications of AI errors, encouraging group investigation and research on similar cases. It emphasizes the impact of these biases on security and health, guiding students to discuss and brainstorm solutions such as improving data quality, increasing diversity in training datasets, and enhancing testing protocols. The resource culminates in preparing a one-minute presentation to summarize their findings and proposed solutions.