According to The Rundown AI on X, Meta open-sourced TRIBE v2, a model trained on 500+ hours of fMRI data from 700+ participants that predicts activity across roughly 70,000 brain voxels in a zero-shot setting, meaning it generalizes to people it never scanned; The Rundown AI also reports the model’s simulated signals are cleaner than raw fMRI because scans contain artifacts like heartbeat, head motion, and machine noise. As reported by The Rundown AI, the approach suggests immediate opportunities for AI-driven neuromarketing tests, rapid cognitive state tagging, and scalable benchmarking for brain computer interface research without bespoke data collection. According to The Rundown AI, the public release positions Meta’s TRIBE v2 as a potential foundation model for multimodal neuroscience tasks, enabling developers to build APIs for content-to-brain response prediction, privacy-preserving user studies, and adaptive media personalization.

According to AI at Meta on X, Meta introduced TRIBE v2, a trimodal brain encoder foundation model trained to predict human brain responses to almost any sight or sound using 500+ hours of fMRI from 700+ participants (source: AI at Meta). According to Meta’s announcement page, the model builds on its Algonauts 2025 award-winning architecture to create a digital twin of neural activity and generalize in zero-shot to new subjects, languages, and tasks (source: go.meta.me/tribe2). As reported by AI at Meta, a public demo is available, signaling practical applications for neuroscience-informed AI, multimodal alignment, and personalized neuroadaptive interfaces in research and healthcare (source: AI at Meta).