Meta’s TRIBE v2: A Foundation Model for Human Brain Prediction
A foundation model that reads the brain, a video generator quietly sunsetted, and a security leak at Anthropic are some of the most striking highlights in AI news this week delivered some of the most striking highlights in AI news this week.
The most technically significant release this week came not from a chatbot or an image generator, but from Meta March 26, 2026, Meta released TRIBE v2, a foundation model designed to predict human neural responses to visual stimuli. On March 26, 2026, Meta released TRIBE v2, a foundation model designed to predict human neural responses to visual stimuli. The model was trained on 500 hours of fMRI data collected from over 700 participants, making it one of the largest and most diverse neural datasets ever used in a publicly released model.
The model was trained on 500 hours of fMRI data collected from over 700 participants, making it one of the largest and most diverse neural datasets ever used in a publicly released model. The model was trained on 500 hours of fMRI data collected from over 700 participants, making it one of the largest and most diverse neural datasets ever used in a publicly released model.
For AI practitioners and researchers, the open release of TRIBE v2 is a significant milestone that enables further exploration of the brain’s inner workings. This enables further exploration of the brain’s inner workings.
For AI practitioners and researchers, the open release of TRIBE v2 is a significant milestone that enables further exploration of the brain’s inner workings. This enables further exploration of the brain’s inner workings.
In a surprising strategic move, OpenAI discontinued its image generation model, Sora, in favor of a new model called Spud. The new model is designed to generate images based on textual descriptions, similar to its predecessor, but with improved capabilities and a focus on generating more realistic images.
In a surprising strategic move, OpenAI discontinued its image generation model, Sora, in favor of a new model called Spud. The new model is designed to generate images based on textual descriptions, similar to its predecessor, but with improved capabilities and a focus on generating more realistic images.
Meanwhile, Anthropic faced a security leak that exposed sensitive information about its AI models. The leak has raised concerns about the security of AI systems and the potential risks associated with their deployment.
Meanwhile, Anthropic faced a security leak that exposed sensitive information about its AI models. The leak has raised concerns about the security of AI systems and the potential risks associated with their deployment.
Leave a Reply