Skip to content
All news

Business · November 19, 2026 · 5 min

Genesis AI raised $500 million on its hands, not its humanoid

The two-year-old robotics startup is now valued at $3.5 billion. Its pitch is a glove worn by workers, a robot hand built to the same shape, and Eno, a wheeled robot that has yet to do a public demo.

Most humanoid fundraises are sold on the body: a walking demo, a dance video, a factory cameo. Genesis AI’s new round was sold on a pair of gloves. On November 17 the company announced a $500 million Series A at a $3.5 billion post-money valuation, co-led by Premji Invest, Khosla Ventures and Eclipse, with NVIDIA, AMD Ventures, Samsung Next, Ericsson, Zoom and Bpifrance among the other backers. That is a big step from the $105 million seed round it announced in July 2025.

Video from Genesis AI on YouTube · Meet Eno

Same glove, same hand

The idea at the center of Genesis AI is simple to say and hard to build. More than 100 workers around the world wear sensor-equipped gloves and head-mounted cameras while doing their normal jobs, and that recording becomes training data for the company’s model, called GENE. The robot hand is built to match the proportions of a human hand and wears the same glove. Genesis calls this a 1:1:1 mapping between the person, the glove and the robot, and the glove captures force, which the company treats as the missing ingredient for handling small, fiddly objects.

The rest of the stack is also built in-house. Genesis World is its simulation platform, used to test skills across many virtual setups before they reach a real robot. The company argues that owning every layer, from motors to model to data, is the only way to get a robot that stays reliable for months instead of minutes.

Eno is still mostly behind the curtain

The robot those pieces are meant to serve is Eno, which looks more like a polished appliance than a metal skeleton: a wheeled base with a tower and two dexterous hands. Genesis says it has built 30 to 40 of them, has finalized the design for mass production, and has picked a contract manufacturer to start building in the US next year. It is piloting Eno with a handful of customers it has not named, and it plans to start in manufacturing, logistics and labs before moving into hospitality and healthcare, then homes and outdoor work.

What is missing is just as telling. TechCrunch’s visit to the company’s Silicon Valley lab saw the model running a pair of off-the-shelf arms that packed, taped and labeled a box. On a second run the arms fumbled a box and struggled to peel off the label. Eno itself was not ready to demonstrate. Genesis also hasn’t said how many Enos it will build next year, and CEO Zhou Xian admitted the dexterous hands currently need replacing after a few weeks of regular work.

A different bet from the rest of the field

Genesis is one of several companies chasing the same prize with very different plans. FieldAI signed a term sheet at a $10 billion valuation to sell one software brain for many kinds of robots, and Runway released a free open-weights robot model for anyone to build on. Genesis is going the other way, keeping the model, the data and the hardware under one roof. Zhou is blunt about where he thinks the value sits: he told TechCrunch the model architecture accounts for “maybe 10%” of the result, and that what matters is the training setup, the data pipeline and the quality of the data.

The glove approach is spreading, too. We covered Robotera selling a robot hand built to match NVIDIA’s data glove earlier this week. Genesis has taken that pairing further by having paid workers collect glove data at scale. Hands are also where the supply chain is thin: Chinese firms shipped 96 percent of dexterous humanoid hands in the first half of 2026, and Zhou named hardware and industrial supply chains as the main bottleneck for his company.

Unlike Figure or Agility Robotics, Genesis isn’t betting on legs. Eno trades walking for wheels and puts its effort into what happens at the fingertips.

“I don’t think we expect or we need another fundamental scientific leap,” Zhou told TechCrunch. “No, we have all the recipes. It’s just about execution.”

Our take

A $3.5 billion valuation for a company with a few dozen robots and unnamed pilots is a bet on the data engine, not on units shipped. The glove idea is sound, and investors clearly agree. But the next 12 months need to show three things the money can’t buy outright: hands that last longer than a few weeks, a named customer using Eno on a real line, and a production number. Follow Genesis AI, compare Eno with other robots on our humanoids page and robot hands guide, and find robotics jobs on our jobs board.

Video: Genesis AI