Alljoined's AIR-O decodes attention from noninvasive EEG in real time
jonxuxu · x · 2026-11-18
Brain-computer interface startup Alljoined introduced AIR-O, a model that decodes what a person is attending to using only noninvasive EEG data, generating an image from scratch in real time depicting the focus of attention. The team calls it the first of a new generation of algorithms that "read the shape of thoughts directly from brain activity."
Context from the repost: Alljoined's founder previously worked on MindEye2 brain decoding at MedARC. The noninvasive + real-time + image generation combination marks notable progress in brain decoding.
More from Research
- Jun-Hua Yong's new series explains how neural nets survive fewer bits — jbhuang0604 · 2026-11-18
- Recursive co-founder Tim Shi on scientific taste, evals, and automating discovery — AnneliesGamble · 2026-11-18
- Research agenda: 'self-inoculation' to keep models' minds intact through RL training — repligate · 2026-11-18
- Robo-ValueRL rates actions by estimated task progress to learn from messy robot data — burkov · 2026-11-18
- RFD4-Proteina protein design GUI now runs in your browser, even on phones — AllThingsApx · 2026-11-18
- DeepBA: a 2M-parameter GNN learns bundle adjustment and beats prior learned SfM from a cold start — Singularitarian · 2026-11-18