Free Intensive Nov 21–22Free 2-day hands-on Vibecoding Intensive on Nov 21–22 at 16:00 MSK (Sat–Sun) · Live Cohort 3 starts Nov 23Join Free →Cohort 3 (Nov 23)
IndustryInsight · On RadarCrunchbase News→ original

Genesis AI Raises $500M at $3.5B Valuation to Teach Robots Through Hand Movements

Genesis AI is teaching robots to use their hands and has raised $500M in a Series A, the first large funding round, at a $3.5B valuation, which is what investors think the whole company is worth. People wearing sensor gloves do ordinary work, while the system records finger movements and touch, and the robot then practices the skill in a simulator. For robotics, the takeaway is simple: recording everyday work may be cheaper than teleoperating a robot.

AI-processed from Crunchbase News; edited by Hamidun News
Genesis AI Raises $500M at $3.5B Valuation to Teach Robots Through Hand Movements
Source: Crunchbase News. Collage: Hamidun News.
Listen to article
Article Fact Sheet & Key Figures
ParameterDetails
Primary sourceCrunchbase News ↗
Editor-in-Chief
Key metric #1Genesis has raised $605M in total since 2024, including $105M in July 2025
Key metric #2The money will go to compute, production of its Eno robots, and hiring; the company has about 100 employees
Key metric #3The company has not disclosed revenue or the number of paying customers

Paris-based startup Genesis AI is teaching robots to work with their hands and has just raised $500M in a Series A round. A Series A is the first major round in which funds buy a stake in a company. A $3.5B valuation means investors believe the entire company is worth that amount, even though it has not disclosed its revenue. The money is backing an idea: robots can learn to handle objects from recordings of how people do it, rather than from expensive teleoperation sessions.

For practitioners, this matters because robotics has long had a bottleneck. Language models learn from enormous volumes of text from the internet. Robotics has no comparable ocean of recordings: you need to know how hands move, how a fragile object is grasped, and what fingers feel on contact. Whoever finds a cheap way to collect this data will have an advantage.

What happened

Genesis AI announced a $500M round at a $3.5B valuation. Several funds led the round, and other investors took part, including Durable Capital Partners. Since its founding in 2024, the company has disclosed $605M in total funding. The previous $105M round was announced in July 2025.

The money will go toward expanding computing infrastructure, increasing robot production, and hiring. The company currently has about 100 employees. Its headquarters are in Paris, and it has an office in San Carlos, California.

The market is heating up as well. According to Crunchbase, global venture investment in robotics reached $9.9B in the third quarter, compared with $4.3B a year earlier, more than doubling. Most of the money goes to companies in China and the US. The adjacent field of physical AI, meaning AI for the physical world, is also attracting heavy investment.

How a robot learns from human hands

Founder Zhou Xian ran into a shortage of training examples while doing his PhD at Carnegie Mellon University. Typically, data is collected by hiring operators who control a robot remotely. This is expensive and slow: the person is not doing their own job but is instead 'driving' a machine.

Genesis proposes a different approach. People put on gloves and work as they normally would, while cameras and sensors record the movements of their wrists and fingers and the tactile signals, meaning what the hand feels on contact. According to the company, its robotic hand is structurally close to the human hand and to the recording glove. That is why movements are easier to transfer from human to robot without complicated recalculation.

The recordings train a model called GENE. Alongside it runs the Genesis World simulator: in a virtual environment, the robot tries a skill many times before attempting it in reality. The company says this makes training faster than relying on physical machines alone. The Genesis project began in 2022 as a research initiative led by Xian together with scientists from more than a dozen universities, building simulation tools. After Xian left university at the end of 2024, he founded the company, which launched in March 2025 and added AI software and its own robots to the simulator.

Robot leasing and open questions

The first robot is called Eno. It is initially aimed at manufacturing, logistics, and laboratories. Later, the company wants to move into hospitality and healthcare, and then into the consumer segment and outdoor work.

The business model is still under discussion. Genesis is considering leasing: a customer pays for the robot and the software as a single service. The company has announced a partnership with one customer and says its sales pipeline includes other clients from the Fortune 500 list. Xian has not disclosed revenue or the number of paying customers. One investor explained the interest by pointing to demonstrations in which robots complete long tasks on their own, rather than showing short clips.

The bet on a 'full stack' (in-house AI, in-house hand, in-house simulator, and in-house testing) is logical, since the reliability of such systems depends on the seams between components. The cost of this approach is high. The company must develop models, hardware, and manufacturing at the same time, which demands a lot of capital and time.

What this means in practice

For those building robots or the data for them, the point is that recording people's ordinary work may turn out to be cheaper than teleoperation, which changes the cost calculation for a dataset. Founders may want to note that money is going to those who show long, autonomous tasks rather than short demos.

Caution is also warranted. The claims about faster training through simulation come from the company itself and have not been independently verified, and there are no commercial metrics. The next verifiable step is the first paying customers for Eno and data on how long the robot operates without human intervention.

Story

Key Questions Answered

What happened?

Genesis AI announced a $500M round at a $3.5B valuation. Several funds led the round, and other investors took part, including Durable Capital Partners. Since its founding in 2024, the company has disclosed $605M in total funding. The previous $105M round was announced in July 2025. The money will go toward expanding computing infrastructure, increasing robot production, and hiring. The company currently has about 100 employees. Its headquarters are in Paris, and it has an office in San Carlos, California.

How a robot learns from human hands?

Founder Zhou Xian ran into a shortage of training examples while doing his PhD at Carnegie Mellon University. Typically, data is collected by hiring operators who control a robot remotely. This is expensive and slow: the person is not doing their own job but is instead 'driving' a machine. Genesis proposes a different approach.

What are the key details on Robot leasing and open questions?

The first robot is called Eno. It is initially aimed at manufacturing, logistics, and laboratories. Later, the company wants to move into hospitality and healthcare, and then into the consumer segment and outdoor work. The business model is still under discussion. Genesis is considering leasing: a customer pays for the robot and the software as a single service. The company has announced a partnership with one customer and says its sales pipeline includes other clients from the Fortune 500 list.

ZK
Zhemal Khamidun — CPO at Alpina Digital, product manager of AlpinaGPT. Building AI products since 2019. Hamidun News is a pet project that grew into a production aggregator with 6,000+ readers.
✦ Hands-On AI Engineering · Hamidun Academy

Build & Ship Production AI Agents Without Boilerplate Code

Start your free 72-hour trial at Hamidun Academy (academy.hamidun.com) — interactive lessons on Claude Code & Cursor, live Tech Radar, and production agent templates.

Start 72h Free Trial →Vibe-Coding Intensive
What do you think?
Loading comments…