Srravya Chandhiramowuli

PhD Student - City, University of London

Data annotation for AI: Working within and around tools, norms and frictions

ChatGPT is the latest of AI systems to make the headlines for its remarkable computational capabilities. Lesser known and rarely acknowledged is the human labours involved in training and supporting these celebrated AI systems. Thousands of workers, particularly in global south regions, create training datasets, validate model outcomes and mimic computational responses to sustain AI’s research, development and use. Yet little is known about what their work entails. What do data labellers do when they label data for AI? Drawing on ethnographic fieldwork conducted at data labelling centres in India, this talk offers insights into the everyday work practices of data labellers, organisational hierarchies, norms, and values that were caught in global flows of resources, rhetoric, and relations of power. We trace these practices, norms and frictions to better understand their influences on everyday annotation work as well as try to answer, why should we, HCI researchers and practitioners, concern ourselves with these seemingly distant realities?

About the speaker:

Srravya Chandhiramowuli is a PhD candidate at the Centre for Human Computer Interaction Design (HCID) at City, University of London. Her current research examines data annotation work practices, paying particular attention to systemic challenges and frictions, to envision and inform just, equitable futures of AI. Srravya’s socio-technical research is shaped by HCI and STS scholarship, as well as her previous experiences in tech policy and software development.