Genesis AI raises $500 million to deploy robots, with reliability still a hurdle
The financing backs an integrated approach to robot intelligence and hardware. A hands-on demonstration exposed the gap between complex tasks and dependable work.
Loading page…
The financing backs an integrated approach to robot intelligence and hardware. A hands-on demonstration exposed the gap between complex tasks and dependable work.
Listen to this story
The Nov. 17 Series A gives Paris-headquartered Genesis AI a $3.5 billion post-money valuation and funds more computing, robot production and hiring. Its approach links sensor-recorded human work, the GENE model, Genesis World simulation and robot hardware, but a repeated packing demonstration faltered and its dexterous hands last only weeks under regular use. Genesis has built 30–40 Eno robots and is piloting them with a handful of undisclosed customers, leaving repeatability and durability as key constraints on expansion.
Premji Invest, Khosla Ventures and Eclipse co-led the round; Nvidia, AMD Ventures and other investors also participated.
More than 100 workers use sensor-equipped gloves and headset cameras to record movements and force data for robot training.
TechCrunch saw the robot complete a box-packing workflow once, then fumble a box and struggle with a label on the repeat attempt.
Genesis AI has raised $500 million to bring its robots into industrial workplaces, but reliable operation remains unfinished business. The November 17 financing values the company at $3.5 billion after the investment. CEO Zhou Xian says Genesis has “cracked dexterous manipulation”; a demonstration for TechCrunch showed both a complex packing workflow and failures when the robot tried it again.
Premji Invest, Khosla Ventures and Eclipse co-led the Series A. Nvidia, AMD Ventures, Bpifrance, Durable Capital Partners and accounts advised by T. Rowe Price Associates also participated, alongside other institutional and individual investors.
The round brings Genesis’s disclosed funding to $605 million, following a $105 million round announced in July 2025. Xian told Crunchbase News that the new money will support more computing infrastructure, increased robot production and additional hiring.
Genesis is developing the robot, its AI model and the systems used to train and test it together. The company’s argument is that reliability requires those pieces to work as one system, rather than treating better software or better hardware as a sufficient answer on its own.
Its training process starts with people doing physical work. Genesis says more than 100 workers around the world wear sensor-equipped gloves and headset cameras. Those devices capture movements and force information, creating examples for the company’s AI model to learn from.
The robotic hand is designed to closely match a human hand and the recording glove. Genesis says that matching structure helps transfer human movements into robot training. Its robots also wear the sensor-equipped gloves, according to TechCrunch.
Those recordings train GENE, the company’s AI model. Genesis World, its simulation platform, provides virtual environments for practicing and testing skills before attempting them with physical machines. Genesis describes a feedback loop in which deployment experience improves the model, simulation and hardware together.
By building it all ourselves, we can obsess over every detail and improve the entire robot faster in one closed loop.
Zhou Xian, Genesis AI co-founder and CEO, in the financing announcement
During TechCrunch’s visit to Genesis’s Silicon Valley facility, the company’s model controlled a pair of off-the-shelf robotic arms. They picked up two small boxes, scanned them and placed them inside a larger box. The arms then folded the outer box closed, taped it shut and peeled a label from printer tape to apply it.
Genesis did not allow reporter Tim Fernholz to rearrange the setup to test how the system would respond. It did agree to repeat the presentation. On that second attempt, the robot fumbled a box and struggled to peel off the small adhesive label.
The demonstration was not performed by Eno, Genesis’s flagship robot. Xian said the model was more capable when operating Eno, but the machine was not ready to demonstrate its abilities during TechCrunch’s visit.
Genesis says it has built 30 to 40 Enos and is piloting them with a handful of undisclosed customers. Its initial targets are manufacturing, logistics and laboratories, with hospitality and healthcare among the industries it plans to pursue later.
Hardware durability is a near-term constraint. Xian told TechCrunch that Genesis’s dexterous hands currently last only a few weeks of regular work before needing replacement. He identified mechanical design and the supply chains needed to produce industrial-grade machines as key bottlenecks.
The company is also exploring how to sell the service. Under a potential rental model, customers would pay for the robot and its software together. Xian told Crunchbase News that Genesis has announced a partnership with LG CNS and has other Fortune 500 prospects, but he declined to disclose revenue or a paying-customer count.
That expansion will build on a team of close to 100 employees, according to Xian. Genesis is headquartered in Paris and also has offices in San Carlos, California.
Loading discussion...
Join the conversation
Explain how maintenance would affect your decision.
Be the first to share a perspective or an experience.
Reader comments
Newest comments first. Replies stay oldest first.