Robot hand company settles Tesla trade secret suit and announces $11M raise

Jay Li, founder of Proception, reflects on the challenges of launching a startup, especially following a lawsuit from Tesla. "I think it’s kind of like a resilience test, or pressure test," he shared in an exclusive interview. Li, previously a technical leader on Tesla's Optimus robot project, faced allegations from his former employer regarding the theft of trade secrets to establish Proception. However, after months of legal disputes, the lawsuit was dismissed earlier this month, allowing Li to focus on a daunting new venture: creating robotic hands that function like human hands. To support this ambitious goal, Proception announced on Monday that it has successfully raised $11 million in seed funding, led by First Round Capital, with additional contributions from Y Combinator and BoxGroup. The startup also revealed that it is now shipping its initial batch of high-dexterity robotic hands to various researchers and robotics firms while opening up for larger orders. Li aims to position Proception as the leading supplier of robotic hands for companies seeking to avoid the lengthy and resource-intensive process of developing what is known in the industry as dextrous manipulation. Despite significant investments flowing into robotics, Li argues that the focus on replicating human-like dexterity in robotic hands has not received enough attention. Tesla CEO Elon Musk has publicly identified the replication of human hand functionality as one of the most significant engineering hurdles in robotics. While Musk suggests that Optimus robots could be operational in factories in the coming years, experts believe that achieving human-equivalent robotic hands remains a challenging endeavor that might take years to realize. Kevin Lynch, director of Northwestern University's Center for Robotics and Biosystems, estimated last year that it could take a decade for robots to perform tasks similar to humans. Li, however, is optimistic that Proception can accelerate this process due to its innovative data collection approach. Most robotic training currently utilizes teleoperators, where humans in virtual reality headsets guide robots without feedback from the objects being manipulated. Proception’s novel solution involves a sensor-laden glove worn by human testers to gather interaction data without requiring a robot to be present. This glove also serves as the sensory interface for Proception's robotic hand, which boasts 22 degrees of freedom and multiple joints per finger for a wide range of movements. Li believes this method allows for more precise, task-specific data collection, ultimately enabling their robotic hands to closely mimic human dexterity. "You need both hardware and data, and those need to come hand-in-hand to get dextrous manipulation to work," Li stated. First Round partner Bill Trenchard, who spearheaded the investment in Proception, expressed confidence in the company’s potential, remarking on the sophistication of their hand and the robust data models backing it. He emphasized that dexterous manipulation is crucial for the future of humanoid robots. Trenchard also commended Li's calm demeanor during the lawsuit, noting that despite the legal challenges, Proception's team remained focused and dedicated. Li remains optimistic about the future, hinting that as Proception grows, Tesla might seek their expertise, stating, "I think it will happen."