Laplace Intelligence, a startup based in Ann Arbor, develops automated red-teaming solutions for robotics and physical AI systems. The company uses high-fidelity simulations to stress-test autonomous vehicles and robots, aiming to identify and resolve potential failures before deployment.
The platform utilizes a post-trained version of Nvidia’s Cosmos simulation environment, which integrates real-world map data and traffic information from providers like TomTom. The system features interactive participants that react to simulated scenarios, allowing for the automated generation of complex edge cases.
Founded a year ago by CEO Howie Sun and chief scientist Henry Liu, the startup originated from research conducted at the University of Michigan. The founders previously published a paper on autonomous driving AI in the journal Nature in 2023.
Laplace currently generates revenue and maintains a profit, with clients including Isuzu, Bot Auto, and Nvidia. The company is also collaborating with T-Mobile on a smart city pilot in San Jose and is exploring applications for warehouse humanoid robotics.
The startup is participating in the Startup Battlefield competition at TechCrunch Disrupt in San Francisco. Laplace aims to raise funds and scale its operations to compete with established players in the sector, such as Applied Intuition.
