Meta has introduced Brain2Qwerty, an advanced AI system that converts brain activity into text with remarkable accuracy. Developed in collaboration with the Basque Center on Cognition, Brain and Language, the system utilizes magnetoencephalography (MEG) and electroencephalography (EEG) to capture brain signals.
These recordings are processed by a deep learning model trained to recognize patterns in brain activity linked to specific keystrokes. The technology achieved up to 80% accuracy in translating thoughts into text, marking a significant milestone in brain-computer interface (BCI) research.
The AI-driven system captures neural activity through an MEG scanner, which takes 1,000 snapshots of the brain per second. This enables the technology to detect the precise moments when thoughts form words and letters.
The AI then maps this data to typed text, generating an on-screen representation of a user’s thoughts. While EEG signals are also used, MEG-based recordings offer much greater accuracy, making this approach a breakthrough in non-invasive brain communication.
Despite its impressive capabilities, Brain2Qwerty faces several challenges. The primary limitation is the requirement for a massive MEG scanner, which weighs half a ton and costs around $2 million.
Additionally, the system demands a magnetically shielded environment to ensure accurate readings, restricting its use to specialized labs. Even slight head movements can impact performance, highlighting the need for further refinements before this technology becomes practical for everyday use.
Looking ahead, Brain2Qwerty’s potential applications in healthcare are immense. It could revolutionize communication for individuals with speech impairments caused by neurological conditions such as locked-in syndrome.
Furthermore, its insights into brain activity and language processing could contribute to future advancements in AI-powered assistive devices. While widespread adoption remains a distant goal, Meta’s investment in BCI research demonstrates a commitment to integrating AI with neuroscience.
