Morning Edition · Sunday, July 12, 2026Published at 1:29 AM EDT · New York
Meta's Brain2Qwerty v2 Decodes Typed Sentences From Brain Scans at 61% Word Accuracy
The non-invasive pipeline reads magnetoencephalography signals, up from 8% for prior surgery-free methods, but still needs a shielded room and a stationary 306-sensor scanner.
Meta's fundamental AI research group detailed Brain2Qwerty v2, a system that decodes typed sentences from non-invasive magnetoencephalography (MEG, a method that measures the magnetic fields produced by neural activity) with no surgery. The…
Continue the AI Intelligence Brief
Track frontier labs, chips, export controls, model releases, regulation, and AI infrastructure.
- 5 AI intelligence signals a day
- Frontier labs, compute, and chips
- Model releases and AI infrastructure
- Source-grounded analysis with confidence labels
The Global Intelligence Brief stays free.
Part of a tracked trend
Non-Invasive Neural Decoding
AI labs increasingly apply machine learning to decode language from non-invasive brain signals, trading fidelity for accessibility and pushing neurotechnology toward broader assistive use.
More from this edition
- Apple Sues OpenAI, Alleging Former Engineers Stole Hardware Secrets for Its ChatGPT Devices
- Meta Starts Charging for Its Own Model With Muse Spark 1.1 and a New API
- Anthropic Returns Claude Fable 5 to Service and Proposes a Shared Jailbreak-Severity Scale
- Cloudflare Splits AI Crawler Controls Into Search, Agent, and Training
- Meta Ships Muse Image and Previews Muse Video, Its First In-House Visual Models
- Musk and Altman Trade Barbs on X as Both Ship New Flagship Models
- SAM-MT Targets Real-Time Segmentation of Many Objects at Once in Video
- Anthropic Publishes the Origin Story of Claude Code, Its Coding Agent
- Anthropic Invites the Public's Hardest Questions About AI
- Anthropic Expands Its Economic Research on How AI Reshapes Work
- Google Consolidates Its Geospatial Models Under Earth AI