The Coded Gaze: Bias in Artificial Intelligence

Joy Buolamwini discusses algorithmic bias in AI, focusing on facial recognition systems. She introduces the concept of the coded gaze, highlighting biases in technology shaped by those in power. Her research, including the Gender Shades study, reveals significant disparities in AI accuracy across gender and skin type. Buolamwini emphasizes the importance of intersectional analysis and ethical AI development, advocating for transparency and responsibility through initiatives like the Safe Face Pledge.

What role does media representation play in the development of biased AI systems?

What changes did companies like IBM and Microsoft make after the Gender Shades study was released?

In what ways can AI be abused even if it is accurate?

What is the Safe Face Pledge and what does it aim to achieve?