Search results for EEG
AI & Enterprise
KAIST develops AI that reads intentions from brainwaves without words
Researchers have developed AI technology that identifies a person’s intentions through brainwaves and adjusts its own behaviour without spoken or physical cues. KAIST said the team led by Sang Wan Lee developed a brain-computer interface technique called Neural Value Alignment with Microsoft Research Asia. The system distinguishes different brainwave signals linked to goal errors and action-method errors, then uses deep learning to interpret EEG in real time and revise AI behaviour accordingly.
AI & Enterprise
ETRI leads international standard for lossless biosignal compression to support medical AI era
A lossless biosignal compression technology developed by South Korean researchers has been adopted as an international standard. The Electronics and Telecommunications Research Institute (ETRI) said its technology was incorporated into a biosignal and general waveform compression standard jointly pursued by ITU-T and MPEG. The standard was approved as ITU-T Recommendation T.261 and MPEG-D Part 8, with ETRI contributing 2 core compression technologies and serving as an editor of the standard document.
Industry
Meta unveils noninvasive BCI that turns thoughts into text without surgery, reaches 61 percent accuracy
Meta has unveiled Brain2Qwerty v2, a noninvasive AI-based brain-computer interface that converts brain activity into text without brain surgery. The system uses helmet-shaped magnetoencephalography equipment and an end-to-end model to reconstruct intended sentences, and it fine-tunes a large language model for neural data. Meta said average word accuracy reached 61 percent. Nine volunteers took part, and results and training code were released alongside a paper in Nature Neuroscience.
