Neuralink participants have now generated over 50,000 hours of unlabeled neural data for BCI pretraining
XFreeze · x · 2026-10-03
According to a post by XFreeze, Neuralink participants have collectively generated more than 50,000 hours of unlabeled neural data.
The company is using thousands of hours of brain activity from each participant to pretrain neural encoders before cursor-control tasks. The goal is to make BCIs faster, more stable, and far less reliant on calibration over time.
This marks a shift from decoding individual brain signals toward building a scalable interface between the human brain and computers.
Related event: Neuralink Pretrains BCI Foundation Model on 50,000 Hours of Neural Data(6 posts)→
More from Embodied
- Phone and robot build a real-time depth map together using Auki spatial computing — broodsugar · 2026-10-03
- Red Dead Redemption 2 hits 70-80 fps on Apple Silicon via GameToMac — ID_R_McGregor · 2026-10-03
- New survey maps 237 methods for embedding physics priors in robot learning — Jan_R_Peters · 2026-10-03
- Scale AI's Alexandr Wang Wows at Stack-chan Robot Showing Squishy Muse Gadgets Character — alexandr_wang · 2026-10-03
- Alexandr Wang Shows Muse Gadget Running on DHH's Omarchy Linux Phone — alexandr_wang · 2026-10-03
- AMD runs MolmoAct2 robot policy locally on a single Ryzen AI Max+ 395 mini PC — DJiafei · 2026-10-03