Genesis AI, a global physical AI research lab and full-stack robotics company, emerged from stealth today with a mission to unlock unlimited physical labor.
The company is building a universal robotics foundation model (RFM) and a horizontal robotics platform, raising $105 million co-led by Eclipse and Khosla Ventures, with participation from Bpifrance, HSG, and visionary leaders Eric Schmidt and Xavier Niel.
Physical labor contributes an estimated $30-40 trillion to the Global GDP, yet over 95 percent of it remains unautomated due to the limitations of current automation solutions. Today’s robotic systems, such as industrial arms, rely on brittle, rigid, and overfitted software stacks.
These systems are narrow in scope, costly to deploy, and challenging to scale. Genesis aims to revolutionize the next generation of general-purpose robots by unlocking unprecedented robustness, flexibility, and cost efficiency – ultimately automating all physical labor.
Genesis brings a data-centric, full-stack approach to physical AI – building a scalable and universal data engine that unifies high-fidelity physics simulation, multimodal generative modeling, and large-scale real robot data collection.
Its simulation stack, developed entirely in-house, will generate rich synthetic data at scale, together with a more efficient and scalable real-world data collection system.
This dual engine of synthetic and real data bridges historically siloed domains to collect the largest-scale, most diverse, and highest quality data to train RFMs.
Zhou Xian, CEO of Genesis, says: “General-purpose robots powered by physical AI will define the next major chapter of human history. While digital AI has made extraordinary progress, physical AI – the intelligence that allows machines to perceive, understand, and interact with the real world – has lagged behind.
“We’re here to change that. By building on the foundations laid by existing digital AI models, we’re bringing human-level intelligence into the physical world.
“Genesis’s unique approach by fueling digital AI knowledge to drive the emergence of physical AI will deliver unmatched capability, scalability, and cost-efficiency to unlock unlimited physical labor. With 75 percent of global companies struggling to fill jobs, physical AI is more essential than ever.”
Founded by top academic and industry technical talents from Mistral AI, Nvidia, Google, CMU, MIT, Stanford, Columbia and UMD, with deep expertise across the full stack of physics simulation, graphics, robotics, and large-scale AI model training and deployment, Genesis is well-positioned to rapidly execute its vision through a differentiated approach to physical AI.
The company also plans to open-source components of its data engine and foundation model to empower developers, researchers, and partners to build on its breakthroughs and accelerate progress across the broader field of physical AI.
Charly Mwangi, Eclipse partner, says: “Even in the most ‘automated’ industries today, the robot-to-human ratio rarely exceeds 1:30, due to the long tail of tasks requiring dexterity, cognition, mobility, and real-world reasoning that current robots simply can’t handle.
“General-purpose robotics is the breakthrough we’ve been waiting for and stands to impact trillions in labor value across sectors. Genesis has the vision, strategy, and world-class team to define the era of physical AI in order to unlock unlimited physical labor through general-purpose robotics.”
Kanu Gulati, of Khosla Ventures, says: “Physical AI has yet to scale like LLMs because collecting and aligning real-world data can be operationally complex.
“Genesis is taking a full-stack approach by integrating best-in-class simulation data with real-world robotics data in a continuous, closed-loop system.
“Owning the entire data pipeline in-house gives them a unique data advantage. We’re excited to back Genesis early as they work to build a universal foundation model for robotics.”
