The Agent RaceJune 30, 2026via MarkTechPost

Meta AI Releases Brain2Qwerty v2: A Non-Invasive MEG Brain-to-Text Pipeline Decoding Typed Sentences at 61% Word Accuracy

Why it matters

Meta's Brain2Qwerty v2 represents a significant capability milestone in brain-computer interfaces using non-invasive MEG, with open-source code release accelerating industry adoption of neural decoding as an emerging AI modality.

Key signals

  • 61% word accuracy on typed sentence decoding
  • Non-invasive MEG (magnetoencephalography) based
  • Open training code released
  • v2 iteration indicates active development cycle
  • Published June 2026

The hook

61% accuracy. Meta just open-sourced a non-invasive brain-to-text decoder that could reshape how we think about AI interfaces.

Meta AI releases Brain2Qwerty v2, a non-invasive MEG brain-to-text pipeline reaching 61% word accuracy with open training code. The post Meta AI Releases Brain2Qwerty v2: A Non-Invasive MEG Brain-to-Text Pipeline Decoding Typed Sentences at 61% Word Accuracy appeared first on MarkTechPost.

The week's key stories, every Friday.

For practitioners and enthusiasts — free, in your inbox.

Free forever. No spam.