Brain-computer interfaces, also known as brain-machine interfaces (BMIs), enable users to control external devices using their thoughts. Since brain surgery is not required, noninvasive brain-computer interfaces may have broader market potential beyond medical uses and could extend into consumer, wearables, gaming, and more. In this study, the UCLA researchers address this performance issue using artificial intelligence for its pattern-recognition capabilities to decode the noisy data from the noninvasive brain-computer interface. They created two AI copilots to help the brain-computer interface users control a robotic arm and a computer cursor. Moreover, the researchers report that the paralyzed patient would not have been able to perform the tasks without the aid of the AI copilot.