Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Division
Browse By Session Type
Search Tips
Personal Schedule
Sign In
Using insights from debates over the meaning and practice of intersectionality (Crenshaw, 1991; Robertson, 2013, 2017), this paper critically examines a burgeoning subfield in computer science focused on fairness, accountability, and transparency in machine learning (FATML). The paper argues that although the subfield has helped highlight the dangers of “bad data” and “bad algorithms,” it remains preoccupied with a set of techno-legal solutions that accept the inevitability of data-driven technologies, on the one hand, and that instantiate a liberal conception of equality, on the other. As a result, the subfield of FATML risks foreclosing a broader conversation about technology and power and consolidating techno-legal expertise with potentially exclusionary and marginalizing effects. Based on the Our Data Bodies (ODB) Project--a qualitative and participatory study of the impact of data collection on marginalized communities in the United States, the paper offers an inclusive approach to collective problem identification and problem solving that begins with the assessment of data-driven technologies and their relation to interlocking systems of social, economic, and racial injustice.