Physical AI Innovations: Pronto Leads the Way in Wearable Camera Technology

Physical AI has taken a significant leap with Pronto’s initiative to enhance its experiment involving wearable cameras in residential spaces. This pioneering step highlights the potential for gathering real-world data, crucial for training AI and robotics systems. In contrast, larger home service competitors like Snabbit and Urban Company’s InstaHelp have chosen not to pursue similar paths.

Pronto’s Pioneering Approach in Physical AI

With an impressive $45 million funding raised this year, Pronto is enlarging its opt-in Pronto Verified service. This service records service sessions while utilising anonymised videos for AI training purposes. As various firms seek methods to amass real-world data for AI and robotics, there’s a noticeable increase in investor interest across sectors like robotics, autonomous systems, and AI-enhanced hardware.

Year-on-Year Funding Trends in Physical AI

According to Startup Superb’s analysis, the Indian Physical AI ecosystem has raised approximately $155 million through 31 deals in 2026 thus far. This figure signifies a rise from $130 million across 25 deals in 2025, $124 million over 24 agreements in 2024, and $91 million from 21 deals in 2023, as per Venture Intelligence data. This funding dataset encompasses startups based in India, alongside foreign firms that are significantly engaged in Physical AI product development or data collection activities in India. However, companies that merely possess engineering or support teams in India or whose Physical AI initiatives are still in an experimental stage, like Pronto, are not included in this funding tally.

Funding Performance in 2026

This year’s funding reflects a 19% increase compared to the total amassed in 2025 and over a 70% rise compared to the levels recorded in 2023. Most of the companies securing funding this year are in the early stages, although substantial amounts are concentrated in a few larger investment rounds. Notably, Innefu Labs, Unbox Robotics, and Deccan AI collectively raised $83 million, making up more than half of the total $155 million raised so far.

Growth-Stage Deals in Physical AI

Innefu Labs spearheaded the segment with a Series B funding round of $30 million, spearheaded by Panthera Growth Partners. This national security-oriented AI enterprise is also venturing into Physical AI and robotics for defence applications. Likewise, Unbox Robotics, a startup concentrating on supply chain robotics, raised $28 million in a round spearheaded by ICICI Venture, with involvement from multiple venture capital firms.

Early-Stage Startups Making Waves

Rekise Marine, a marine robotics company, secured $9.7 million in a seed funding round co-led by Accel and NKSquared, engaging Indian shipyards in their vessel design processes. Meanwhile, Human Archive, based in San Francisco and Bengaluru, acquired $8.2 million in a round led by Wing Venture Capital. This Y Combinator-supported venture aimed to create large-scale human sensorimotor datasets for training robots and Physical AI systems.

Investment Trends and Major Rounds

In addition, Bengaluru’s SwitchOn recorded $8 million in a pre-Series B funding round led by IvyCap Ventures, focusing on AI-powered vision systems for automated quality checks in manufacturing. Other notable funding rounds included Spector.ai’s $6.7 million led by IvyCap Ventures and ANSCER Robotics’ $5.4 million Series A led by IAN Alpha Fund. Meanwhile, Mowito raised $3 million in a pre-seed round aimed at developing AI foundation models for industrial robotic arms, and Flo Mobility garnered $2.5 million in a pre-Series A round.

Emerging Data Infrastructure for Physical AI

A distinct group of firms is concentrating on gathering, annotating, and training infrastructure essential for Physical AI systems. Human Archive is engaged in establishing human sensorimotor datasets, while Neocambrian AI has created a dedicated robotics data factory within India.

Importance of Real-World Data in Physical AI Development

It is crucial that Physical AI models have access to real-world data that accurately depicts human actions and interactions. Much of this data is specifically gathered for robotics and embodied AI scenarios instead of being obtained from pre-existing internet datasets. Pronto serves as another case study demonstrating how such data could be produced, showcasing how organisations with employees in real-world settings might become valuable contributors to training data.

Potential for Larger Funding Rounds

Skylark Labs, founded by Amarjot Singh, is significantly active within India and collaborates closely with the country’s defence sectors. The firm is in negotiations to secure between $100 million and $150 million, potentially elevating its valuation to over $1 billion. This funding round would surpass any prior individual Physical AI funding round involving a company related to India in 2026.

Challenges in the Physical AI Landscape

Companies building hardware in the Physical AI sector must navigate higher deployment costs and extended commercial timelines. Privacy issues and obtaining consent will remain crucial considerations, especially regarding data collected in residential environments.