Genesis AI Emerges with $105M Funding to Develop General-Purpose AI Models for Robots Using Synthetic Data

Genesis AI, a startup focused on creating a foundational AI model to power a variety of robots, has recently emerged from stealth mode with a significant $105 million seed round co-led by Eclipse Ventures and Khosla Ventures. Founded last December by Zhou Xian, a Ph.D. in robotics from Carnegie Mellon University, and Théophile Gervet, a former research scientist at the French AI lab Mistral, the company aims to develop a general-purpose model that can automate a broad spectrum of repetitive tasks, ranging from laboratory work to household chores. Unlike large language models, which are trained on extensive text datasets, AI models for robotics require data from the physical world. Collecting this real-world data is often expensive and time-consuming. To address this challenge, Genesis is leveraging synthetic data generated by a proprietary physics engine designed to accurately simulate the physical world. This approach allows the company to develop AI models more efficiently and potentially at a lower cost compared to traditional methods. The genesis of the company's synthetic data engine dates back to an academic project led by Xian, involving researchers from 18 universities. Many of these researchers have now joined Genesis, contributing to a team of over 20 specialists in robotics, machine learning, and graphics. Kanu Gulati, a partner at Khosla Ventures, expressed confidence in the team's capabilities, saying, "It’s a big unknown: Will anybody have a large robotics foundation model that will generalize across tasks? That’s a bet we want to go after. Of all the teams we have seen, we like Genesis’s approach for going after robotics foundation models." Competitors in this space include Physical Intelligence, which recently secured a $400 million round, and Skild AI, valued at $4 billion earlier this year. Both companies are also working on general-purpose AI models for robotics, but Genesis believes its proprietary simulation engine gives it a competitive edge over platforms like NVIDIA’s. Currently, Genesis is operating out of two offices, one in Silicon Valley and the other in Paris. The company's near-term goal is to release its foundational model to the robotics community by the end of the year, marking a significant milestone in its journey to revolutionize robotic automation. This substantial investment underscores the increasing interest and financial backing in AI-driven robotics, reflecting the industry's belief in the potential of such technology to transform various sectors through enhanced automation and efficiency.