Many research and industry organizations outsource labor-intensive processes of data generation, annotation, and algorithmic verification—or data work—to workers worldwide through digital platforms. A subset of the gig economy, these platforms consider workers independent users with no employment rights, pay them per task, and control them with automated algorithmic managers. This talk focuses on the data worker population in Venezuela—a Latin American country experiencing a socioeconomic crisis exacerbated by the COVID-19 pandemic and where a significant number of data workers in the Global South are located—to understand the coloniality of data work, or the historical patterns of power that define contemporary data production. This talk will look at how power imbalances are reproduced in outsourced data work in a context in which platform companies profit from marginalized workers in a socioeconomic condition characterized by a lack of regulations and protections on the population. The data production process ensures that workers’ voices are suppressed in annotation tasks through algorithmic management and surveillance, resulting in datasets generated exclusively by clients, with their worldviews encoded in algorithms through training. The coloniality of data work is thus characterized by an extractivist method of generating data that privileges profit and the epistemic dominance of those in power. This talk argues that data production is a process and that the quality of data and artificial intelligence algorithms depend on incorporating a plurality of workers’ voices and the decommodification of their labor away from market dependency.
Julian Posada is a Postdoctoral Associate and incoming Assistant Professor of American Studies at Yale University. His research integrates theories and methods from information studies, sociology, and human–computer interaction to study technology and society. I am currently researching the relationship between human labor and data production in the artificial intelligence industry. This project centers on the experiences of outsourced workers in Latin America employed by digital platforms to produce machine learning data and verify algorithmic outputs.