New AI Translates Brain Activity into Text Without Surgical Implants A groundbreaking artificial intelligence system called the semantic decoder has been developed, capable of translating brain activity into continuous text. This system has the potential to revolutionize communication for individuals who are unable to speak due to conditions like stroke. Unlike other language decoding systems, the semantic decoder is noninvasive and does not require surgical implants. Using an fMRI scanner, the system captures the essence of a person's thoughts, albeit imperfectly, about half of the time. The researchers from The University of Texas at Austin, led by Jerry Tang and Alex Huth, utilized a transformer model similar to those used by Open AI's ChatGPT and Google's Bard. Participants in the study listened to podcasts in the scanner, and their brain activity was measured to generate corresponding text. Although the system is not yet practical for use outside the laboratory due to its reliance on fMRI machines, the researchers believe it could be adapted to more portable brain-imaging systems such as fNIRS. The study addressed concerns about misuse and emphasized the importance of consent and participant cooperation. The researchers also demonstrated the system's ability to describe events from silent videos based on brain activity. The work was supported by foundations such as the Whitehall Foundation and the Alfred P. Sloan Foundation. Co-authors of the study include Amanda LeBel and Shailee Jain, and the researchers have filed a patent application related to this work. #BrainActivity #TextTranslation #ArtificialIntelligence #SemanticDecoder #Noninvasive #StrokeCommunication #UniversityOfTexas #TransformingThoughts #fMRI #PortableBrainImaging #JerryTang #AlexHuth #OpenAI #ChatGPT #GoogleBard #PodcastTraining #EthicalTechnology #ConsentIsKey #SilentVideoDescription #GroundbreakingResearch #WhitehallFoundation #SloanFoundation #PatentApplication