Automation
Genesis AI raises $500M for wheeled humanoid robot Eno
Paris startup's Series A brings valuation to $3.5 billion as it pursues general-purpose robotics with human-hand design and custom foundation model.
Key takeaways
- Genesis AI raised $500 million at a $3.5 billion valuation from investors including Nvidia, AMD, Khosla Ventures and former Google CEO Eric Schmidt.
- The company's Eno robot uses a wheeled base and human-modeled hands with 20 degrees of freedom, enabling training on human demonstration data rather than robot-specific footage.
- Genesis's GENE foundation model can perform multi-step tasks including playing piano, cooking and solving Rubik's Cubes, targeting both industrial and residential deployment.
- The funding will expand AI research teams, robotics engineering staff and training data collection operations as Genesis scales production.
- Genesis's approach of mimicking human hand anatomy addresses a core challenge in robotics AI training by dramatically expanding available training datasets.
Genesis AI secures massive Series A for general-purpose robotics
Paris-based Genesis AI SAS has closed a $500 million Series A funding round at a $3.5 billion valuation, according to details first reported by SiliconANGLE. The round was led by Premji Invest, Khosla Ventures and Eclipse, with participation from the venture arms of Nvidia, AMD, Zoom, Samsung and Ericsson. Former Google CEO Eric Schmidt and other technology executives also contributed to the round.
The capital will fund expansion of the company's AI research team, robotics engineering staff, and training data collection operations as Genesis works to scale deployment of its flagship robot.
Wheeled design with human-inspired hands
Genesis's robot, called Eno, takes a different approach from the humanoid robots currently drawing attention in the industry. Rather than walking on legs, Eno uses a wheeled base that can adjust its height through a hinged mechanism consisting of two large panels. This configuration allows the robot to adapt to different tasks while maintaining mobility.
The distinguishing feature of Eno's design is its hands. Genesis built robotic hands with 20 degrees of freedom—compared to 27 in human hands—and covered them with a soft, skin-like material. The hands are deliberately modeled after human anatomy, which solves a critical training data problem.
Traditional robotics AI models require training datasets featuring the specific robot design they'll control. A four-legged robot needs footage of four-legged robots performing tasks. For specialized designs, collecting sufficient training data becomes prohibitively difficult.
By mimicking human hand structure, Genesis can train Eno's AI model on demonstrations of humans performing tasks with their hands. This approach dramatically expands the available training data pool and simplifies the data collection process.
Custom foundation model demonstrates versatility
Eno is powered by GENE, a custom foundation model that Genesis says is easier to train than conventional robotics-optimized neural networks. The company has released demonstration videos showing GENE controlling Eno to play piano, cook meals and solve a Rubik's Cube—tasks that require dozens of sequential steps.
According to Genesis co-founder and CEO Zhou Xian, this versatility positions Eno for deployment in both industrial settings and residential environments. "We're working to build the world's best general-purpose robot," Xian said. "By building it all ourselves, we can obsess over every detail and improve the entire robot faster in one closed loop."
Why it matters
The $3.5 billion valuation and investor roster signal growing confidence that general-purpose robotics can move beyond factory automation into broader commercial and consumer applications. Genesis's approach of using human demonstration data rather than robot-specific footage could accelerate AI model development across the robotics industry, potentially lowering barriers to entry for companies building specialized robotic systems. The participation of major semiconductor and communications companies also suggests strategic interest in the infrastructure requirements for scaled robot deployment.
The funding details were first reported by SiliconANGLE.