The global AI training data market is projected to reach $25 billion by 2030, with Physical AI validation identified as a technically demanding and rapidly growing segment where Humyn Labs aims to play a key infrastructure role.

Humyn Labs commits USD 20 million to scale its human intelligence layer for Physical AI, expanding data collection and validation infrastructure for real-world datasets.

As an Indian physical AI data startup, Humyn Labs is investing to extend data collection across global markets, including Southeast Asia and Latin America, to support diverse, real-world datasets.

The initiative emphasizes voice data and multilingual datasets, with development in 33 languages to reach underrepresented markets.

Humyn Labs focuses on egocentric, first-person data collection that captures real-world human activity and movements in commercial, agricultural, and residential settings to build context-rich datasets.

The data effort aims to show how people see, navigate, and interact with their surroundings, producing training-ready datasets for Physical AI across real-world contexts.

Data collection will expand to include egocentric visuals and movements from diverse environments to train physical AI systems.

Data sourcing will be decentralized across the Global South, leveraging a network of about 15–18 customers and targeting scalable contracts in the $10–$15 million range per engagement.

Expansion targets include India, Southeast Asia, and Latin America to create geographically diverse datasets and improve AI adaptability globally.

This move reflects a broader industry trend toward investing in data infrastructure to gain a competitive edge in AI development, underscoring the value of high-quality, real-world data.

Co-founders emphasize continuous human data as foundational for real-world AI systems and scalability, with a focus on the Global South as a driver of AI’s future.

Gupta highlights the Global South as a core site for AI data creation and future development, not merely a source.