Human Archive pays Indian gig workers to wear camera hats and sensor devices, collecting real-world physical training data for AI and robotics labs
The Data Collection Model
Human Archive, a startup founded by researchers from the University of California, Berkeley and Stanford University, is paying gig workers in India to wear camera-equipped caps and sensor devices. The workers are tasked with capturing real-world physical training data that artificial intelligence and robotics labs are racing to acquire. The startup’s approach relies on a distributed workforce, compensating individuals for recording daily activities and movements in natural environments.
Technology and Equipment
The camera-equipped caps worn by the gig workers contain compact recording modules, while the sensor devices track motion, orientation, and other physical parameters. This hardware setup is designed to generate high-fidelity training datasets that can teach AI models and robotic systems how to navigate, interact with, and manipulate objects in the physical world. The collected data includes visual feeds from the head-mounted cameras and synchronized sensor readings, providing a multi-modal view of human actions and environmental context.
Research and Development Background
The founding team at Human Archive brings together expertise from two leading academic institutions. The researchers previously worked on computer vision, robotics, and machine learning projects at UC Berkeley and Stanford, where they recognized a growing bottleneck in the AI industry: the scarcity of diverse, real-world physical training data. Many robotics labs and AI companies struggle to obtain large-scale, labeled datasets that accurately represent human behavior in uncontrolled settings, as opposed to synthetic or lab-generated data.
Operational Details and Worker Engagement
Human Archive recruits gig workers through digital platforms in India, offering payment for each data collection session. Workers are instructed to wear the camera caps and sensors during routine activities such as walking, carrying objects, or performing household tasks. The startup emphasizes that participation is voluntary and compensated, with workers retaining no ownership over the recorded data. The exact payment rates and number of workers deployed have not been disclosed by the company.
Market Context
Bitcoin is trading at $75,972, down 2.1% in the last 24 hours. Ethereum is at $2,073.74, down 2.6% in the same period. Data as of May 26, 2026.