Stanford’s Wu Tsai Neurosciences Institute and the Institute for Human-Centered Artificial Intelligence awarded a second round of joint grants to five cross-disciplinary teams aiming to advance both brain science and machine intelligence. The projects span efforts to refine depression subtypes by linking EEG and fMRI data; model infant learning by fusing babies’ brain scans with first-person video; build a foundation model to decode how genetically similar neurons diverge into distinct cell types; map how psychotherapy dialogue influences deep brain dynamics using large language models and topology; and align AI odor representations with human neural activity and subjective experience.
Leaders of the initiative framed the awards as a bid to accelerate ethically grounded, human-inspired AI while informing clinical practice. The program, launched in 2022, underscores a growing convergence between neuroscience and AI, with teams targeting near-term impacts in mental health treatment and longer-term advances in data-efficient, experience-aware AI systems.
Related articles:
— Resting-state connectivity biomarkers define neurophysiological subtypes of depression
— Allen Cell Types Database
— Ego4D: An egocentric video dataset for daily life activities
— The Human Connectome Project: Mapping human brain circuits
