Meta has unveiled Brain2Qwerty v2, a new artificial intelligence system that can decode brain activity into text without requiring surgical implants, according to a company announcement. The system represents a significant advancement in brain-computer interface research, achieving word accuracy levels previously attainable only through invasive techniques.
Breakthrough in non-invasive brain decoding
Brain2Qwerty v2 is described by Meta as the highest-performing end-to-end pipeline capable of real-time sentence decoding from non-invasive brain recordings. The company stated that the technology is approaching levels of accuracy previously exclusive to techniques requiring brain surgery, such as stereotactic electroencephalography and electrocorticography.
Unlike these invasive methods, Brain2Qwerty uses non-invasive recordings to decode intended text. Meta trained the system using approximately 22,000 sentences collected from nine volunteer participants. Each participant spent about 10 hours wearing a magnetoencephalography (MEG) device while actively typing.
How Brain2Qwerty v2 works
The system employs end-to-end deep learning to decode language directly from raw brain signals, rather than relying on manually designed pipelines to identify neural events. Meta explained that large language models were fine-tuned on neural data, enabling the system to use semantic context to bridge the gap between noisy brain signals and coherent language. The company also deployed AI agents to explore optimizations in the decoding process, with final training configurations selected by engineers.
Performance and accuracy improvements
Meta reported that Brain2Qwerty v2 achieved a word accuracy rate of 61 percent, a significant improvement over the 8 percent word accuracy achieved by other non-invasive methods. For the best-performing participant in the study, the system reached a word accuracy rate of 78 percent, with more than half of all decoded sentences containing one word error or less.
The company also found that decoding accuracy improved as more training data became available, suggesting that the performance gap between non-invasive and surgical approaches could be reduced further through larger datasets.
Potential impact and future research
Meta stated that the research could help millions of people suffering from brain lesions and other conditions that prevent them from communicating. To support further research, Meta announced that it is releasing the full training code for Brain2Qwerty v1 and v2. Its research partner, the Basque Center on Cognition, Brain and Language (BCBL), will also release the Brain2Qwerty v1 dataset.
Meta said the work forms part of its broader efforts to develop open foundational models of the brain and advance neuroscience research aimed at improving the diagnosis, treatment and understanding of neurological disorders. The company highlighted its TribeV2 model for perception encoding, NeuralSet for processing brain data at scale, and NeuralBench for systematically evaluating brain models. These initiatives form part of Meta's Digital Brain Project, under which it recently announced a USD 5 million fund to support the creation of open neuroscience datasets.
