Meta AI releases non-invasive BCI Brain2Qwerty v2

// 1h agoRESEARCH PAPER

Meta AI releases non-invasive BCI Brain2Qwerty v2

Meta AI has released Brain2Qwerty v2, a non-invasive brain-computer interface (BCI) system that decodes brain activity into real-time sentences using deep learning and large language models. The system achieves an average word accuracy rate of 61% from magnetoencephalography (MEG) recordings.

// ANALYSIS

Combining LLMs with neural recordings is a massive win for non-invasive BCIs, but the reliance on room-sized MEG scanners means consumer-grade portable applications are still far off.

* The system reaches unprecedented non-invasive word accuracy (61% average), narrowing the gap with surgical implants.

* While MEG performance is strong, the EEG setup still suffers from a high 67% character error rate, highlighting the difficulty of deploying this on portable consumer wearables.

* Meta's decision to release the code and dataset will accelerate neuro-AI collaboration but raises necessary long-term questions around mental privacy and neurorights.

// TAGS

brain-computer-interfacebcimagnetoencephalographymegdeep-learningmeta-airesearch

DISCOVERED

1h ago

2026-06-29

PUBLISHED

1h ago

2026-06-29

RELEVANCE

8/ 10

AUTHOR

omarsar0

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

VIDEO44m ago

Netlify Agent Runners Auto-Provision Postgres and Auth

Netlify released a video demonstration of Agent Runners, showing how the built-in AI coding agents can provision a PostgreSQL database, configure user authentication/identity, and set up project skills to make an application production-ready in minutes. Because Agent Runners execute within the Netlify dashboard ecosystem, they operate with full project context, including environment variables, database connections, and deployment settings.

RESEARCH52m ago

PhysisForcing injects physics into video simulation

Generative video models like DiTs are increasingly used as embodied world simulators but often suffer from physically implausible artifacts such as discontinuous trajectories and object deformation. PhysisForcing addresses these issues by focusing supervision on physics-informative regions using trajectory and relational alignment losses, significantly improving physical consistency and closed-loop robotic planning success.

UPDATE1h ago

Cognition launches Devin Fusion hybrid-model harness

Cognition has introduced Devin Fusion, a hybrid-model harness that dynamically routes tasks mid-session between a sidekick agent and a frontier model. This approach reduces the cost of frontier-level intelligence by 35% while maintaining performance on complex coding tasks.