The gig workers who are training humanoid robots at home

# Summary: Gig Workers Training Humanoid Robots from Home

Gig workers in remote locations, including medical student Zeus in Nigeria, are being hired to record themselves performing everyday tasks while wearing iPhone headsets. These recordings capture human movement and behavior, which are then used to train humanoid robots through machine learning algorithms. The work represents a new form of distributed labor where individuals can earn income by simply documenting their daily activities.

This trend reflects a broader shift in how companies develop AI-powered robots. Rather than hiring specialized mocap studios or expensive motion-capture facilities, companies are leveraging a global pool of gig workers to generate training data at minimal cost. This approach allows companies to scale data collection rapidly while providing income opportunities to workers in regions where such opportunities are limited.

The development raises important questions about labor practices, data privacy, and the economics of AI training. While gig workers gain flexible income sources, concerns exist about fair compensation, worker protections, and ownership of movement data. As humanoid robots become increasingly prevalent across industries, the workers training them from home studios represent both an emerging labor market and a case study in how AI development intersects with global inequality.