Y Combinator-backed Human Archive just raised $8.2 million to record tens of thousands of hours of Indian blue-collar worker data. This funding comes amidst intense scrutiny after reports of home services startups like Snabbit using gig workers to train physical AI systems. The ethical questions around workers unknowingly training algorithms that could replace their jobs are now center stage.
How We Got Here
The San Francisco and Bengaluru-based startup was founded by 20-year-old UC Berkeley and Stanford dropouts who quickly garnered YC backing. Earlier this week, Pronto and Snabbit faced public debate for their pilots using worker tracking data to train AI models, putting Human Archive in the spotlight.
The Numbers
- The $8.2 million round was led by Wing Venture Capital (an early Snowflake backer) and NVP Capital.
- Angel investors include executives from OpenAI, NVIDIA, Google, Meta, and DoorDash AI Research.
- Human Archive provides workers with hardware rigs featuring downward-facing 4K cameras, depth sensors, and wide-angle lenses to capture hand movements.
- The collected data undergoes anonymisation, annotation, and processing before being sold to frontier labs and robotics companies.
- Founders are Rushil Agarwal, Raj Patel, Samay Maini, and Shloke Patel.
What Happens Next
🇮🇳 Why This Matters for India
For the 50 lakh gig workers in Mumbai, Delhi, and Hyderabad, this model presents a double-edged sword: a potential short-term income source against long-term job displacement.
The Take
Human Archive represents a frontier where ethical considerations lag significantly behind technological capabilities. The real losers are India's blue-collar workers who lack the agency to decide how their labor fuels algorithms that could render them redundant.
Source:
Inc42 ↗