Paralyzed Man Controls Robotic Arm for Record Seven Months Using AI Brain Implant

A groundbreaking brain-computer interface adapts to shifting neural activity, enabling long-term, precise robotic arm control through thought alone.

Researchers at UC San Francisco developed an AI-powered brain-computer interface (BCI) that allowed a paralyzed man to control a robotic arm for seven months, a significant improvement over previous systems that lasted only days.
The system uses AI to adapt to natural shifts in brain activity, maintaining stable and precise control over time without frequent recalibration.
The participant, paralyzed by a stroke, was able to perform complex tasks such as picking up blocks, opening a cabinet, and filling a cup with water using the robotic arm.
The study included a training phase where the participant practiced with a virtual robotic arm, refining his control before transitioning to the real-world device.
Researchers are now working to improve the system for smoother and faster movements and plan to test its application in home environments to enhance the independence of individuals with paralysis.