The global triumph of Chinese AI depends on a hidden workforce of data annotators in the Global South. This comparative study examines the lived experiences of 38 workers in Egypt, Morocco, and Kenya who label images, transcribe speech, and refine datasets for Chinese tech companies through cross-border chains. Drawing on semi-structured interviews analyzed through reflexive thematic analysis, four themes emerged: (1) a layered architecture of subcontracting and deliberate client opacity; (2) a granular system of algorithmic discipline and gamified precarity; (3) a piece-rate economy delivering volume yet precarious earnings; and (4) regulatory voids filled by local intermediaries and emerging worker solidarity, including organizations such as Kenya’s Data Labelers Association. Engaging with frameworks of planetary labor markets and algorithmic imposition, the study shows that Chinese firms’ exclusive BPO model creates discipline fundamentally different from Western platform precarity, while national contexts dramatically reshape outcomes across the three countries.
AnwarM. A. (2025). Value chains of AI: Data training firms, platforms, and workers. In The future of labour (pp. 198–213). Routledge. https://doi.org/10.31235/osf.io/cjn63
BridiM. A.ZanoniA. P. (2024). Plataformas digitales de microtrabajo en brasil. Estudios Sociológicos, 42(Dossier issue), 1–19. https://doi.org/10.24201/es.2024v42.e2693
ChandhiramowuliS.TaylorA.HeitlingerS.WangD. (2024). Making data work count. Proceedings of the ACM on Human-Computer Interaction, 8(CSCW1), 1–26. https://doi.org/10.1145/3637367
9.
ChaudhuriB.ChandhiramowuliS. (2024). Tracing the displacement of data work in AI. Engaging Science, Technology, and Society, 10(1–2), 8–31. https://doi.org/10.17351/ests2024.2983
10.
ChenS. (2016). The materialist circuits and the quest for environmental justice in ICT’s global expansion. tripleC: Communication, Capitalism & Critique, 14(1), 121–131. https://doi.org/10.31269/triplec.v14i1.695
11.
ChenS.ZhaoY. (2022). Ecological civilization: A blindspot in global media coverage of China’s environmental governance. Environmental Communication, 16(2), 195–208. https://doi.org/10.1080/17524032.2021.1981419
12.
CorbinJ. M.StraussA. (1990). Grounded theory research: Procedures, canons, and evaluative criteria. Qualitative Sociology, 13(1), 3–21. https://doi.org/10.1007/bf00988593
13.
FerrazD. B.CostaM. M. M. D. (2024). Microtrabalho na inteligência artificial. Revista Brasileira de Sociologia do Direito, 11(2), 25–43. https://doi.org/10.21910/rbsd.v11i2.807
14.
FuP.LinJ. (2026). From planetary to state embedded AI stacks, the re-territorialisation of China’s data annotation industry. Big Data and Society, 13(1), 20539517261421470. https://doi.org/10.1177/20539517261421470
15.
FuP.LinZ.WangW. Y. (2025). Operationalizing AI governance, data annotation, La Qi and manual alignment in China. Information, Communication and Society, 29(6), 1–23. https://doi.org/10.1080/1369118x.2025.2526162
HowsonK.JohnstonH.ColeM.FerrariF.Ustek-SpildaF.GrahamM. (2022). Unpaid labour and territorial extraction in digital value networks. Global Networks. https://doi.org/10.1111/glob.12407
HwangJ. (2024). The gig economy and its implications for labor laws and worker protections. International Journal of Science and Research Archive, 13(1), 3405–3416. https://doi.org/10.30574/ijsra.2024.13.1.1714
JohnstonH.SilbermanM. S.HowsonK.WoodcockJ. (2025). How can working conditions for online crowdworkers be improved?Work in the Global Economy, 5(2), 261–281. https://doi.org/10.1332/27324176y2025d000000037
MaryalaB. T. R. (2025). Global ethical AI data standard. World Journal of Advanced Engineering Technology and Sciences, 15(2), 2836–2841. https://doi.org/10.30574/wjaets.2025.15.2.0851
28.
MiceliM.PosadaJ. (2022). The data-production dispositif. Proceedings of the ACM on Human-Computer Interaction, 6(CSCW2), 1–37. https://doi.org/10.1145/3555561
29.
MiceliM.SchuesslerM.YangT. (2020). Between subjectivity and imposition, power dynamics in data annotation for computer vision. Proceedings of the ACM on Human Computer Interaction, 4(CSCW2), 1–25. https://doi.org/10.1145/3415186
Potocka-SionekN. (2022a). Crowdwork and global supply chains. In A research agenda for the gig economy and society (pp. 215–234). Edward Elgar Publishing. https://doi.org/10.4337/9781800883512.00019
44.
Potocka-SionekN. (2022b). Niewidzialni pracownicy, czyli kto stoi za sztuczną inteligencją. Studia z Zakresu Prawa Pracy i Polityki Społecznej, 29(2), 105–119. https://doi.org/10.4467/25444654spp.22.010.15684
SantosH.OliveiraS. S. (2025). No centro do furacão da heteromação. Revista Convergência Crítica, 1(21).
48.
SchmidtF. A. (2022). The planetary stacking order of multilayered crowd AI systems. In GrahamM.FerrariF. (Eds.), Digital work in the planetary market (pp. 137–156). The MIT Press. https://doi.org/10.7551/mitpress/13835.003.0012
TubaroP.CasilliA. A.CornetM.Le LudecC.Torres CierpeJ. (2025). Where does AI come from? A global case study across Europe, Africa, and Latin America. New Political Economy, 30(3), 359–372.
51.
TubaroP.CasilliA. A.CovilleM. (2020). The trainer, the verifier, the imitator. Big Data and Society, 7(1), Article 2053951720919776.
WangD.PrabhatS.SambasivanN. (2022). Whose AI dream, in search of the aspiration in data annotation. In CHI 22 (pp. 1–16). ACM. Article 582. https://doi.org/10.1145/3491102.3502121
WinquesK.MagnoloT. S. (2024). Como a ia enxerga seus trabalhadores?Revista Latinoamericana de Ciencias de la Comunicación, 23(46). https://doi.org/10.55738/alaic.v23i46.1143
YangT.MiceliM. (2024). “Guilds” as worker empowerment and control in a Chinese data work platform. Proceedings of the ACM on Human-Computer Interaction, 8(CSCW2), 1–27. https://doi.org/10.1145/3686904
