Artificial intelligence algorithms read human thoughts

This AI algorithm uses a helmet with sensors, its task is to analyze specific electrical activity in the human brain when thinking, then convert those thoughts into text.

Scientists say the new device will help care for patients who are mute due to stroke or paralysis while also allowing for seamless control of devices such as bionic limbs and robots, helping humans move give instructions just by thinking about them.

Lead researcher Professor CT Lin said the new device marks a significant breakthrough in decoding raw EEG waves directly into language, opening new frontiers in neuroscience and AI.

Previously, scientists had to perform surgery to implant electrodes into the human brain or scan with an MRI machine to decode brain signals into language. This technology is very complex, high cost and difficult to use in life. As for new technology, just a simple helmet on the head can read a person's thoughts.

Test results on 29 people show that the content decoded by AI almost perfectly matches human thoughts.

The accuracy of decoded text is currently around 40%. And scientists are continuing to research to improve equipment to increase that number to 90%.