Fight AI Bias With Bias
Explore the concept of combating AI bias through strategic bias implementation in this 47-minute conference talk from Strange Loop 2022. Delve into how human biases influence AI model outputs and how AI amplifies these biases, leading to socio-technical harms that disproportionately affect legally protected classes in the United States. Learn about techniques to de-bias datasets and build intelligent systems that are fair, equitable, and trustworthy. Gain insights from Noble Ackerson, a former Google Developers Expert and Responsible AI specialist, as he expands on the topic of bias and its implications for AI adoption, safety, and well-being. Discover practical approaches to implementing data governance privacy principles, eliminating bias, and promoting fairness in machine learning applications across commercial and federal sectors.