Humyn Labs, a startup that supplies real-world footage of people performing everyday tasks so robots can learn from it, has raised $47 million through a mix of equity and debt.
The funding, reported by Inc42, arrives as interest in physical AI, the branch of artificial intelligence focused on machines that act in the real world, continues to draw investor attention. It also comes amid open questions about privacy, consent and the effect of such data on workers.
How The $47 Million Round Breaks Down
Of the total, $26 million came as equity in a round led by Accel. The remaining $21 million was raised as debt from Lighthouse Canton and Trifecta Capital.
Blending equity with venture debt allows a fast-growing company to fund expansion while limiting how much ownership it gives up, a structure that has become more common among startups with meaningful revenue.
Where The Money Is Going
Humyn Labs plans to use the capital to expand its robot deployment operations into Brazil, South Korea and India. It also intends to upgrade its data collection hardware and strengthen its processing capabilities.
Part of the funding will support a larger presence in San Francisco, London and Seoul, cities that sit close to major robotics developers, AI labs and potential customers.
A Fast Start For A Young Company
The company was founded earlier this year and says it reached $50 million in annual recurring revenue within six months of operating. That figure is the startup's own claim.
Rapid growth of that kind, combined with rising demand for data to train physical AI systems, appears to have helped attract global backers to the round.
Cameras, Labelling And A Global Footprint
Humyn Labs has built its own two-camera recording hardware to capture people carrying out routine activities. Its software then processes and labels the footage, turning it into training material for robots.
Its data collection network now covers more than 20 countries across Latin America and Asia. Robots need large volumes of varied, real-world examples to learn physical tasks, and that kind of data is far harder to gather than the text and images used to train chatbots.
Privacy, Consent And A Labour Paradox
The business model raises significant questions. Recording inside homes and factories can capture personal information, which makes strong consent processes and secure storage essential.
There is also a tension at the heart of the work: people are recording their own movements to train machines that may eventually automate those same jobs. How Humyn Labs handles privacy protections, data usage and fair pay for contributors is likely to draw scrutiny as it grows.
What The Deal Signals For Physical AI
The round shows that investors see data collection as a key bottleneck in robotics, and are willing to fund companies that can gather it at scale across many countries.
For founders, it highlights an opportunity in the infrastructure behind physical AI rather than only in the robots themselves. It also shows that companies in this space will need clear answers on consent, data ownership and worker compensation to sustain trust as they expand.
Source: Inc42





