Mimic Raises $16M to Train Robots in Human Dexterity for Industrial Uses

• Mimic has raised $16 million to scale its AI model and robotic hand for complex industrial tasks.
• The system is already in pilot use with Fortune 500 manufacturers and global logistics firms.

Mimic, a robotics company based in Zurich, has raised $16 million in seed funding to grow its physical AI technology. The round was led by Elaia, with backing from Speedinvest and others. The company is building a general-purpose AI model and robotic hand to automate dexterous, manual tasks in manufacturing, assembly, and logistics.

Mimic trains its foundation AI model using movement data from skilled workers wearing custom tracking gear while doing their normal factory work. This data is used to teach robots through imitation learning, allowing them to perform tasks that require human-like dexterity. The AI powers Mimic’s robotic hand, which pairs with off-the-shelf robot arms to manipulate objects, recover from errors, and adapt to changes in real time. The system is already being piloted with Fortune 500 manufacturers and global logistics providers.

“Humanoids are exciting, but there aren’t many industrial scenarios where the full-body form factor truly adds value,” says Stephan-Daniel Gravert, co-founder and CPO at mimic, in an official press release. “Our approach pairs AI-driven dexterous robotic hands with proven, off-the-shelf robot arms to deliver the same capabilities in a way that is much simpler, more reliable and rapidly deployable. 

The funding will support the development of Mimic’s AI model and robotic hand, and help expand deployments with industrial partners. The company was recently selected for the AWS Generative AI Accelerator, a program for startups applying advanced AI to real-world problems.

🌀 Tom’s Take:

Mimic is building dexterity from the ground up, using real human motion data to train models that actually work on the factory floor.

Source: Globenewswire / mimic