Zeus turns on his ring light and straps his iPhone to his forehead in his studio apartment, raising his hands in front of him like a sleepwalker to perform recorded tasks. The process captures human movement data for training home-based humanoid robots according to the primary source (https://www.technologyreview.com/2026/04/01/1134863/humanoid-data-training-gig-economy-2026-breakthrough-technology/).

The report identifies this as part of gig-economy data collection occurring in multiple locations to support physical AI development. Similar recording methods have been documented in other robotics training programs per the cited article.

Primary source coverage from MIT Technology Review on April 1 2026 states the practice occurs after regular work hours using consumer devices.