Machines uncover human thoughts | AI “mind reading” becomes possible!

The research on brain-computer interaction has made great progress in the past decade. This technology application scene is described in the movie “The Matrix”, William Gibson’s novel “Neural Roamer”, and the game “Sword Art Online”. To put it simply, “brain-computer interaction” is the transmission of information between the human brain and the computer to meet human needs and purposes. Now, a group of computer scientists and neuroscientists from the University of San Francisco have taken this technology one step further.

Dr. Joseph Margin and his colleagues used two neural networks to decode volunteer brain activity and determine what they were talking about. They realized the real-time translation of neural activity to text, and were able to obtain a fairly high degree of accuracy.

A total of four volunteers participated in the experiment, and 250 electrodes were implanted in the brains of each participant to monitor changes in brain waves. They read aloud between 30 and 50 sentences, and the research team tracked their neural activity through electrode arrays as they spoke, and then input the data into machine learning algorithms.