• 2 min read
BrainCo shows mind-controlled robots at WAIC
BrainCo unveiled a brain-to-robot platform at WAIC in Shanghai, translating EEG signals into robot commands in under 200 milliseconds.

Image: iXBT
BrainCo has unveiled a platform that lets people control robots with their thoughts. At the World Artificial Intelligence Conference (WAIC) in Shanghai, the Chinese company demonstrated its Brain-Controlled Robot AI Platform, where a user wears an EEG headset, forms an intention mentally, and a robot carries out the command without buttons, voice control, or physical movement.
The system works in three stages. First, the electroencephalography headset reads the user’s signals. Then, AI identifies the person’s intention — such as “pick up an object” — and converts it into commands for the robot. According to BrainCo, the full cycle from brain signal to robot action takes less than 200 milliseconds.
During the demonstration, a robotic arm controlled by a person’s thoughts picked up a cup and lifted an apple. BrainCo says the platform is compatible with different kinds of off-the-shelf robots, including humanoids, manipulators, and quadruped robots. That means it can be used in existing robotics labs without requiring custom hardware.
BrainCo also presented a data collection system for training embodied AI models, which operate in the physical world through robots. The company says one of the field’s main bottlenecks is a lack of high-quality data: teaching a machine to fold clothes or handle fragile objects requires large sets of real-world examples.
Its new system uses a mobile rig with two robotic arms and a sensor glove that records human movements while also capturing brain signals during each action.

Recommended reading
Nvidia’s deepfake detector scans video in 22 ms
BrainCo has worked on brain-computer interfaces (BCI) since 2015 and previously focused on medical rehabilitation technologies, including prosthetics controlled by nerve and muscle signals. This latest project brings that experience into robotics, where the company is developing both a brain-based control interface and AI models for physical machines.
AI Editor
Ava covers the rapidly evolving world of artificial intelligence, from foundational models and research labs to the real-world economics of intelligence. With a background in computational linguistics, she cuts through the hype to find out what actually works. She firmly believes that benchmarks are just marketing until reproduced in the wild.
via iXBT


