Genesis AI Advances Dexterous Robotics with Integrated Hardware and AI
Genesis AI, a startup developing robots that learn complex physical tasks from human movement, has secured $500 million to scale production of its Eno robot and expand its training infrastructure, aiming to bridge the gap between lab demonstrations and reliable industrial automat
Genesis AI, a company focused on enabling robots to master complex physical tasks by observing human movements, has announced a significant $500 million Series A funding round. This capital infusion, co-led by Khosla Ventures, Premji Invest, and Eclipse, will be directed towards expanding the company's computing infrastructure, scaling robot production, and growing its team of nearly 100 employees. The funding signals strong backing for Genesis AI's approach to developing integrated robotic systems capable of dexterous manipulation in industrial settings.
The core of Genesis AI's offering is its vertically integrated strategy, encompassing both robot hardware and the AI software that controls it. The company's method involves recording human hand and finger movements, along with touch information, using cameras and sensor-enabled gloves. This data is then used to train its proprietary AI model, GENE, which powers its flagship humanoid robot, Eno. This approach aims to overcome the historical challenge of acquiring sufficient, high-quality training data for robot learning.
A crucial component of Genesis AI's training pipeline is its simulation software, Genesis World. This platform allows robots to practice and refine skills in virtual environments before deploying them in the physical world, a process the company claims significantly accelerates development compared to relying solely on physical machines. Co-founder and CEO Zhou Xian emphasized the necessity of building an entire technology stack to ensure tight integration and deliver a reliable product, stating that it "requires the intelligence, hardware and everything that goes into training and testing the robots to work as one system."
Genesis AI's robots are designed with hands that closely mimic human proportions, facilitating the transfer of learned movements from human data. TechCrunch observed a demonstration where Genesis AI's physical AI model, operating off-the-shelf robotic arms, performed a complex packaging workflow, including scanning, boxing, sealing, and labeling. While the demonstration showed a hiccup with a dropped box and fumbled label, a Genesis executive noted this "hiccup data" is valuable for improving the model's performance in edge cases.
Initially targeting manufacturing, logistics, and laboratories, Genesis AI's Eno robot is currently in production ramp-up for industrial customers. The company is exploring a rental model, offering the robot and its software as a combined service, and has already announced a partnership with LG CNS. Looking ahead, Genesis AI plans to expand its reach into sectors such as hospitality, healthcare, and eventually consumer and outdoor applications, indicating a broad vision for generalized robotic intelligence.
Zhou Xian, who started a robotics research collaboration named Genesis in 2022 during his Ph.D. at Carnegie Mellon University, co-founded the company in 2024 with Théophile Gervet. The company formally launched in March 2025, expanding from simulation tools to include AI software and robot hardware. Xian believes that the current bottleneck for truly generalized robotic intelligence lies in hardware capability and supply chains, rather than fundamental scientific breakthroughs, stating, "We have all the recipes. It’s just about execution."
Sources
- 01 Genesis AI Raises $500M At $3.5B Valuation To Bring Its Robots To Factories And Warehouses — Crunchbase News
- 02 Genesis AI raises $500 million to deploy glove-wearing robots — TechCrunch — Insider