Ghostcrafting AI: Under the Rug of Platform Labor

Platform laborers play an indispensable yet hidden role in building and sustaining AI systems. Drawing on an eight-month ethnography of Bangladesh's platform labor industry and inspired by Gray and Suri, we conceptualize Ghostcrafting AI to describe how workers materially enable AI while remaining invisible or erased from recognition. Workers pursue platform labor as a path to prestige and mobility but sustain themselves through resourceful, situated learning - renting cyber-cafe computers, copying gig templates, following tutorials in unfamiliar languages, and relying on peer networks. At the same time, they face exploitative wages, unreliable payments, biased algorithms, and governance structures that make their labor precarious and invisible. To cope, they develop tactical repertoires such as identity masking, bypassing platform fees, and pirated tools. These practices reveal both AI's dependency on ghostcrafted labor and the urgent need for design, policy, and governance interventions that ensure fairness, recognition, and sustainability in platform futures.