Human Archive raised $8.2 million to pay India's gig workers to wear camera-equipped caps while cleaning homes, cooking meals, and performing other physical tasks, capturing first-person video data the startup sells to robotics labs training physical AI models. The Silicon Valley startup has 1,000 active headsets deployed across India and pays workers a base rate of $1 per hour, well below the $2.63 to $4.20 competitors offer but profitable due to its on-the-ground operational presence.
Wing Venture Capital led the round with participation from NVP Capital, Y Combinator, and angels from OpenAI, Nvidia, Google, Meta, and multiple robotics research labs. The funding comes as robotics companies and frontier AI labs face a critical shortage of real-world training data showing humans performing everyday physical work, a bottleneck for models that must operate in unstructured environments.
Human Archive differentiates itself by capturing synchronized data across multiple sensors beyond standard video. The startup deploys 50 different devices including tactile gloves, full-body motion capture suits, wrist cameras, and RGB-D headsets that pair color imagery with real-time depth information. The company operates 7 different hardware products it switches between depending on task modality.
“To capture data, we started with iPhones; then we built our own custom rigs and caps. Now we have more than seven different hardware products that we use interchangeably across different modalities. After data collection from different devices, we worked on synchronizing data from all these different sources.”— Raj Patel, Human Archive CEO
Key facts
- 01Human Archive raised $8.2M from Wing Venture Capital, NVP Capital, Y Combinator, and angels from OpenAI, Nvidia, Google, and Meta.
- 02The startup has 1,000 active headsets deployed across India's gig economy, recording workers performing everyday tasks.
- 03Workers earn $1 per hour base rate for wearing camera caps, compared to $2.63 to $4.20 per hour paid by competitors.
- 04Human Archive deploys 50 different devices across 7 hardware products to capture synchronized video, depth, motion, and tactile data.
- 05India's Ministry of Electronics and Information Technology is investigating consent mechanisms for egocentric data collection startups.
The company partners with smaller home services and cloud kitchen startups in India to offer customers discounted services in exchange for consenting to data collection. When a worker arrives, customers choose between a discounted price with recording or full price without. The startup says disputes over service quality are common and recordings help resolve them, driving customer opt-in rates.
Larger Indian gig-economy platforms rejected partnership proposals from Human Archive. Urban Company CEO Abhiraj Singh Bhal stated publicly the company would not participate in such arrangements. Co-founder Rushil Agarwal posted that Pronto founder Anjali Sardana called him stupid when he proposed a data partnership. The rejections forced Human Archive to work with smaller, less established service providers.
Human Archive is training internal AI models on its collected data and testing them on physical robots to demonstrate dataset quality to potential customers. The company says every major lab and university has expressed interest in running experiments on the data due to the novelty of synchronized multi-sensor capture at scale. A public dataset release is planned.
Privacy concerns surround the data collection program. Human Archive says it complies with India's Digital Personal Data Protection Act by displaying privacy notices and consent information, anonymizing all data, and blurring faces from recordings. India's Ministry of Electronics and Information Technology is investigating consent mechanisms and data-collection practices of startups gathering egocentric data through home service workers.
The company is expanding beyond India into Southeast Asia and the U.S., building a platform for anyone globally to participate in data collection and earn money. Early-stage pilots in the U.S. aim to offer customers services like cleaning or cooking in exchange for data collection by participating workers, replicating the Indian model in a higher-wage market.
Human Archive competes in a market where multiple well-funded startups are racing to build physical AI systems. The company's success depends on whether it can secure enough partnerships to collect data at the volume and diversity AI labs require, and whether its multi-sensor synchronization approach commands a pricing premium over simpler video-only datasets. The $1 per hour wage keeps costs low but raises questions about whether the startup can scale globally at that rate.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
