AI’s Human-in-the-Loop Paradigm: Corporate-Framed ‘Collaboration’ Masks Structural Job Displacement & Power Consolidation
Original framing: “Mira Murati Wants Her AI to ‘Keep Humans in the Loop’” — Wired
Research in labor economics (e.g., Acemoglu & Restrepo, 2018) shows that automation displaces workers not by replacing entire jobs but by fragmenting tasks, leading to precarious employment. Studies on gig work (e.g., De Stefano, 2016) demonstrate that ‘human-in-the-loop’ systems often result in lower wages, lack of benefits, and psychological distress due to algorithmic management. Neuroscientific evidence suggests that repetitive microtasking degrades cognitive function over time, contradicting the narrative that such work is ‘collaborative’ or ‘skilled.’
Murati’s ‘humans-in-the-loop’ AI is not a radical departure but a continuation of capitalism’s long tradition of reducing human labor to commodified inputs, from Taylorism to the gig economy.
The framing obscures how this model deepens global inequality, particularly for marginalized workers in the Global South, while entrenching corporate control over AI development. Historical parallels reveal that ‘collaboration’ in such systems is a euphemism for precarity, a trickster’s inversion of mutual benefit. Cross-culturally, alternatives like Indigenous AI sovereignty or worker cooperatives demonstrate that AI can serve collective well-being rather than extractive profit. The path forward requires dismantling the illusion of ‘ethical AI’ within corporate frames and replacing it with democratic, community-owned models that prioritize human dignity over machine optimization. Without this, ‘keeping humans in the loop’ will merely mean keeping them in their place—trapped in a loop of exploitation designed by those who profit from their precarity.