In brief
- Meta introduced Brain2Qwerty v2, a non-invasive AI system that decodes brain activity into text.
- The model achieved 61% average word accuracy, compared with about 8% for previous non-invasive methods.
- Meta released the training code for Brain2Qwerty v1 and v2, while its research partner is releasing the v1 dataset.
Meta on Monday introduced Brain2Qwerty v2, an AI system that translates brain activity into text using non-invasive brain recordings. The company said the research is intended to help people who have lost the ability to communicate because of brain lesions.
The system records brain activity using a helmet-like magnetoencephalography (MEG) scanner, a non-invasive brain imaging device commonly used in neuroscience research. It then feeds those raw neural signals into an end-to-end AI model that reconstructs the sentences a person is trying to type. Meta said it further improves accuracy by fine-tuning large language models on neural data, allowing the system to use semantic context when interpreting noisy brain recordings.
“We trained Brain2Qwerty v2 on approximately 22,000 sentences from nine volunteer participants, each recorded for 10 hours wearing a magnetoencephalography (MEG) device while actively typing,” Meta wrote. “Instead of relying on hand-crafted pipelines to detect neural events, we use end-to-end deep learning to decode directly from raw brain signals.”
Meta said Brain2Qwerty achieved a 61% average word accuracy, compared with roughly 8% for previous non-invasive methods. The company is releasing the system's code and dataset as part of its Digital Brain Project, which also includes a $5 million fund to support open neuroscience datasets.
Meta also said decoding accuracy improved as the amount of training data increased, suggesting additional data could further improve performance. The company said AI agents explored possible optimizations for the decoding pipeline before engineers selected the final training configuration.
In an accompanying paper published in Nature Neuroscience, Meta researchers argued that while AI has significantly improved brain-to-text decoding, most high-performing brain-computer interfaces still depend on surgically implanted electrodes, making them difficult to scale because of the risks tied to brain surgery and the challenges of maintaining implants over time.
Meta said Brain2Qwerty v2 approaches levels of accuracy previously achieved only with techniques requiring brain surgery. The company said its non-invasive approach could help bridge the gap between invasive neuroprosthetics and communication systems that do not require surgery.
“Our hope is that this work, done in the open, advances neuroscience to identify, diagnose, and treat neurological disorders faster than in siloes,” Meta wrote.
The announcement comes as brain-computer interface research accelerates, including by Elon Musk’s Neuralink and Merge Labs, backed by OpenAI CEO Sam Altman, developing technology to help restore communication for people with neurological disorders.
While companies such as Neuralink and Synchron are pursuing implanted interfaces that require surgery, a growing number of researchers and startups are using AI to improve the performance of non-invasive systems. In September 2024, startup Neurable introduced AI-powered EEG headphones designed to monitor focus and cognitive fatigue. A year later, MIT spinout AlterEgo unveiled a wearable that converts silent neuromuscular signals from the face and throat into text and commands, positioning it as a practical alternative to implanted brain-computer interfaces.
Meta did not immediately respond to a request for comment by Decrypt.