Demystifying bias in data and AI: How can an algorithm be biased?

Data science and artificial intelligence inform decisions in research, business, and politics. Over the last several years, scholars like Safiya Noble and Cathy O'Neil have drawn increasing attention to the ways that data- and algorithm-driven decision making problematically reproduces prejudice and systemic injustice.

In this conversational workshop, we explore several case studies to show how the data that drives decisions can embed social biases, like racism and sexism, and how the algorithms used in research and business can themselves be oppressive. This workshop will build your familiarity with key issues and suggest further readings so that you'll be better equipped to take a critical perspective in relation to our data-driven world.

Political division in social media algorithms, bias in healthcare data, and racism in facial recognition AI for policing are some of the bias-in-technology news stories we saw in 2020. These issues affect us all, but not everyone is an expert on both the technologies and the social issues they pose. This series of one-hour discussions will break down some of the basic technologic concepts and the prominent voices in the criticism of their social impact to give a bird’s eye view of the conversation in and out of academia. We hope to give attendees of all disciplines and expertise a space to learn about and contribute to the emerging conversation by using tangible examples of both the technologies and cases of bias in them along with resources to dig deeper.