Mecka, which collects human motion data to train humanoid robots, raised a $60M Series B led by Sequoia, with participation from Nvidia, M12, and others

Humanoid robot training data now costs less to acquire, potentially accelerating development timelines.

Robot data startup Mecka AI has secured $60 million in a Series B funding round, with Sequoia leading the investment. Nvidia and Microsoft’s venture fund M12 also participated. The company, founded in 2024, focuses on collecting and analyzing human motion data to train humanoid and other types of robots. Mecka AI pays individuals to record themselves performing everyday tasks while wearing body sensors and using smartphones, aiming to provide the essential human-generated data for robotic systems. This move parallels the role data-labeling companies play in training large language models. Mecka AI was reportedly nearing this funding round at a $500 million valuation.

The significant funding round for Mecka AI underscores a growing trend of specialized data collection for AI, extending beyond LLMs into the complex domain of robotics. This investment suggests that the foundational challenges in AI development are shifting towards the physical world, requiring large-scale, nuanced data acquisition strategies. Companies are increasingly looking to replicate human capabilities in robots, necessitating detailed motion and interaction data that startups like Mecka AI are poised to provide.

This influx of capital into robotics data startups indicates a maturing market for AI infrastructure, where the reliable and accurate collection of real-world data is becoming a critical bottleneck. The participation of major tech players like Nvidia and M12 signals their strategic interest in supporting the ecosystems that will power future robotic applications. As more companies enter this space, competition to acquire high-quality human motion data will likely intensify, potentially driving innovation in data collection methods and analysis.