Turing Award winner Yann LeCun participates in investment as Swiss embodied AI startup Enact Intelligence closes a $30 million Pre-Seed funding round
Swiss embodied AI startup Enact Intelligence has completed a $30 million Pre-Seed funding round, co-led by Redalpine and Notion Capital, with participation from Yann LeCun, Arthur Mensch, Thomas Wolf, and others. The company integrates force perception into robot actuators and combines models with a data collection system to enhance dexterous manipulation capabilities; it is developing a foundation model and supporting platform for general-purpose robotic manipulation. The funding will be used to expand its 12-person team fivefold by the end of 2027 and to drive the actual deployment of its first batch of robots with industrial partners.
Body · AI translation
Robot Frontier reported on October 23 that recently, Swiss embodied startup Enact Intelligence ended its stealth operating status and announced the completion of a $30 million (about 200 million yuan) Pre-Seed round of financing, co-led by Redalpine and Notion Capital, with participation from First minute Capital and other institutions.
Angel investors in this round include Mistral co-founder Arthur Mensch, AMI Labs founder and former Meta Chief AI Scientist Yann LeCun, Path Robotics founder Andrew Lonsberry, Hugging Face co-founder Thomas Wolf, and others.
Enact Intelligence was founded in March 2026, headquartered in Zurich, Switzerland, and focuses on R&D of robot force-tactile perception and dexterous manipulation technology.
Most robots mainly rely on visual perception of their surroundings. Enact Intelligence believes that this approach makes it difficult to accurately determine the force the robot itself applies and the force state of objects when grasping, contacting, and manipulating objects, thereby limiting the robot's ability to perform complex tasks.
Based on this, the company adopts a hardware-software collaborative R&D approach, integrating force perception into robot actuators and obtaining force information directly from inside the actuators.
This approach is similar to how humans perceive the force they apply through their muscles, enabling robots not only to recognize their surroundings through vision but also to perceive their own motion state and force state.
On this basis, Enact Intelligence further combines models and data collection systems to help robots understand contact states with objects and improve dexterous manipulation capabilities.
It is understood that Enact Intelligence is developing a foundation model and supporting platform for general robot manipulation, enabling robots of different forms to complete manipulation tasks across different tasks and environments.
Enact Intelligence founder Manuel Yves Galliker previously held positions at 1X Technologies including Senior Robot Control Engineer and Head of Control and Embedded Team, and participated in the development of the humanoid robot NEO. He later served as a technical researcher at Physical Intelligence, mainly participating in VLA model R&D.
▲Enact Intelligence founder Manuel Yves Galliker
The company consists of 12 people, and its core technical personnel come from companies such as Google DeepMind, 1X, Amazon Robotics, Helsing, and Physical Intelligence.
Enact Intelligence plans to use the funds from this round to expand its team to 5 times its current size by the end of 2027, accelerate hardware development and technology R&D, and promote the first batch of robots to carry out actual deployments with industrial partners.
At present, the company has held discussions with several large European industrial enterprises, covering fields such as manufacturing, infrastructure, retail, and food processing.