Cunyan Ma maps AI’s hidden global labor force — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Cunyan Ma created a freely available interactive map linking technology companies and their intermediaries to data workers worldwide, as part of his work with Brown University’s Center for Technological Responsibility, Reimagination and Redesign.
- The map combines three evidence streams: existing data-worker databases, Ma’s own data collection and news reports documenting workers’ conditions.
- Ma said his primary motivation was to visualize “AI colonialism” in an immediately accessible way and fill the gap left by anecdotal journalist reports by showing a larger picture.
- AI models must be trained by humans, with work that can include labeling objects in photos or parts of speech, transcribing long speech passages, rating chatbot responses and identifying tumor boundaries in medical scans.
- Data workers must view horrifying images or text passages to label disturbing material, allowing AI systems to recognize and filter such content.
- Global South data workers are often low-wage human inputs behind algorithmic curation of social-media posts, driving routes and movie choices.
Why it matters: Ma’s free map links technology companies and intermediaries to low-wage data workers in the Global South using three evidence streams: established databases, his own collection and news reports. For those companies and workers, the visible chain places routine annotation and exposure to disturbing content inside the material story of AI, expanding a picture previously limited to anecdotal accounts.
Ask SkimNews
