• 2 min read
BrainCo says thoughts can now steer robots
BrainCo unveiled a non-invasive EEG system in Shanghai that lets people control robots with their thoughts while generating new training data.

Image: Maximalfocus / Unsplash
Chinese company BrainCo has unveiled what it calls the first “brain-to-robot” interface, a system designed to let human operators control robots using their thoughts. Reported by the South China Morning Post, the platform was introduced at the World Artificial Intelligence Conference (WAIC) in Shanghai, described as China’s biggest AI tech event.
The system relies on non-invasive EEG (electroencephalogram) headsets that read neural activity and convert it into instructions for robots. As with other brain-computer interfaces, it depends on large volumes of training data that link specific patterns of brain activity to intended commands, with AI algorithms matching a thought to the correct action.
It’s called the Brain-Controlled Robot AI Platform, and BrainCo says it can be used not only to control robots directly but also to connect with robotic arms and other devices. One example the company gave is using thoughts alone to tell a robotic arm to pick up a cup.
BrainCo also says wider use of the platform will help generate more real-world data, which could improve the underlying models over time. According to the source, access to high-quality training data remains one of the industry’s main bottlenecks.

Recommended reading
Some AI problems stay unsolvable, even with infinite data
“The embodied AI industry has made remarkable progress on what robots can do on their own. We believe the next decisive frontier is about how robots understand the humans they work with.”
The announcement points to a broader push toward physical AI in robotics, where smarter models, better dexterity, and new input methods such as thought control are starting to converge.
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 TechRadar


