Researchers from Meta's AI team in the United States, PSL University in France, and the Rothschild Foundation Hospital have jointly developed a non-invasive method for converting 'brain activity to text' that incorporates AI technology. Using deep learning algorithms, this technology can decode brain activity during keyboard typing based on electroencephalogram (EEG) or magnetoencephalogram (MEG) signals and translate it into text. The relevant research findings have been published in the latest issue of the journal Nature Neuroscience.