MECKA AI RAISES $60M TO TRAIN ROBOTS WITH HUMAN DATA

Mecka AI is building AI models for robotics by leveraging human movement data—gestures, walking patterns, and other physical information—sourced from consumer devices and wearable sensors. The startup's approach addresses a core challenge in robotics: obtaining diverse, high-quality training data. Rather than building proprietary datasets from scratch, Mecka AI taps existing streams of human movement data from body sensors and iPhones to teach robots how humans move and interact with their environment. The $60 million funding round—led by the $25 million Series A—provides capital to scale data collection, expand the AI models, and accelerate product development. The funding reflects investor confidence in the potential for consumer device data to power the next generation of robotic systems. Robotics companies have long struggled with the data bottleneck: training effective robot behaviors typically requires expensive, time-consuming data collection. Mecka AI's model sidesteps some of this friction by repurposing movement data already being generated by millions of smartphone and wearable users. The startup operates in a competitive landscape where other robotics firms are pursuing similar paths—using human demonstrations and motion capture data to train robotic systems. However, Mecka AI's focus on accessible consumer devices as data sources sets it apart from approaches requiring specialized hardware. The company faces practical questions around data privacy, consent, and the extent to which human movement captured in one context translates to effective robot training. As robotics funding accelerates across the industry, success will depend on whether the startup can convert raw movement data into models that produce reliable robotic behavior across diverse tasks and environments.