Meta is approaching to decode human thoughts and read minds thanks to its newly developed artificial intelligence tool. It has been revealed that the artificial intelligence model created by the American technology giant can predict thought-out sentences with an accuracy rate of 80%. Researchers of artificial intelligence at Meta, the parent company of Facebook, are making significant progress in reading human minds and understanding thoughts. Meta’s artificial intelligence model can accurately predict thought-out sentences with an 80% accuracy rate. Moreover, this does not require a chip to be implanted in the brain. Meta’s approach combines Magnetoencephalography (MEG) and Electroencephalography (EEG) technologies. While EEG records the brain’s electrical signals through electrodes placed on the scalp, MEG measures the magnetic fields of synapses with a sort of helmet-like device. The combined use of both technologies ensures more accurate predictions. According to Meta, this method is at least two times more effective than using EEG alone. It was reported that the artificial intelligence model was trained with brain recordings of 35 volunteers. These volunteers had their brain activities recorded while typing specific sentences on a keyboard. In tests on new sentences, it was reported that the artificial intelligence could predict the written characters with an 80% accuracy rate. Artificial intelligence researchers aim not only to convert thoughts into text but also to understand how ideas are expressed in language. Such research was quite challenging for neuroscientists due to the disruption of brain signals caused by mouth and tongue movements. However, this error source is eliminated during the typing process with a keyboard. One of the ultimate goals of this research is to enable individuals who have lost their ability to speak due to accident or illness to communicate solely with their thoughts. However, tests conducted so far have only involved healthy individuals. It is unknown how artificial intelligence will function in individuals with brain damage. Additionally, there is a significant technical obstacle to be overcome: a specially insulated room is required for MEG measurements. Moreover, the devices used are overly large and cumbersome for daily use. The future development of this groundbreaking technology is eagerly anticipated.