Open AccessDISEMBEDDEDNESS IN MACHINE LEARNING DATA WORK
Author(s) -
Julian Posada
Publication year - 2021
Publication title -
selected papers of internet research
Language(s) - English
Resource type - Journals
ISSN - 2162-3317
DOI - 10.5210/spir.v2021i0.12015
Subject(s) - commodity , big data , raw data , politics , work (physics) , profit (economics) , latin americans , artificial intelligence , business , computer science , marketing , economics , engineering , market economy , political science , microeconomics , law , mechanical engineering , programming language , operating system
Firms and research organizations require humans to annotate raw datato make it compatible with machine learning algorithms. These tasks are often outsourced toindividuals worldwide through labor platforms or infrastructures that serve as marketplaceswhere labour is exchanged as a commodity. The firms that operate them consider workers as“independent contractors” without the social and economic benefits of traditional employmentrelations. This presentation explores the personal networks of Latin American data workerswho train and verify data for machine learning algorithms from their homes. A series ofin-depth interviews and an analysis of a self-completion questionnaire and web traffic datasuggests that these workers are embedded of networks of trusts build on online and offlineinteractions. These findings show a continuation of exploitative supply chains in thecurrent artificial intelligence market, where wealthy companies and research institutions inadvanced economies profit from the economic and political situation of developing countriesto access disembedded labor. This paper concludes by arguing that, though outsourced onlinelabour, artificial intelligence developers not only extract value from their workers, butalso indirectly from their communities and personal networks.