# Meta's Brain2Qwerty v2 Decodes Typed Sentences From Brain Scans at 61% Word Accuracy

The non-invasive magnetoencephalography pipeline improves on roughly 8% for prior approaches, with the best participant reaching 78%, and ships with open code and data.

- Published: 2026-07-09T05:32:14.424Z
- Canonical: https://polylog.news/ai/2026-07-09/meta-s-brain2qwerty-v2-decodes-typed-sentences-from-brain-sc
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Meta AI](https://ai.meta.com/blog/brain2qwerty-brain-ai-human-communication/), [MarkTechPost](https://www.marktechpost.com/2026/06/30/meta-ai-releases-brain2qwerty-v2-a-non-invasive-meg-brain-to-text-pipeline-decoding-typed-sentences-at-61-word-accuracy/)

Meta's FAIR group released Brain2Qwerty v2, a non-invasive pipeline that decodes typed sentences from magnetoencephalography (MEG) recordings, a way of measuring brain activity that requires no surgery. The system feeds raw MEG signals into…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-07-09/meta-s-brain2qwerty-v2-decodes-typed-sentences-from-brain-sc (subscription information: https://polylog.news/pricing).