Miceli, MilagrosPosada, Julian2022-04-212022-04-212022-10http://hdl.handle.net/10625/61007This “dispositive” analysis explores the sites where production of machine learning (ML) data is outsourced. It comprises examination of 210 data-work instruction documents, 55 interviews with data workers, managers, and requesters, and participant observation. Findings show that specific discourses encoded in task instructions reproduce and normalize the worldviews of requesters. Precarious working conditions and economic dependency alienates workers, making them obedient to instructions. Outsourced workers are paid as low as a few cents per task, usually lack social protection traditionally tied with employment relations, and are subject to systems of control and surveillance. Tables provide a synopsis of task segmentation with brief analysis.application/pdfenPRECARIOUS EMPLOYMENTARTIFICIAL INTELLIGENCEALGORITHMSMACHINE LEARNINGBASE OF PYRAMIDSEARCH ENGINESDATA COLLECTIONDATA MININGSURVEILLANCELABOUR LAWHUMAN RIGHTS VIOLATIONSSOCIAL PROTECTIONSOUTH AMERICAGLOBALJournal Article