Summary
C-BRAIN is working with NVIDIA on AI tools and computing infrastructure for analysing Alzheimer’s research data, including information held by member institutions. The consortium reports early analysis-time savings and has set longer-term goals for finding new disease mechanisms.
The Consortium for Biomedical Research and AI in Neurodegeneration (C-BRAIN) is working with NVIDIA on AI tools and computing infrastructure for Alzheimer’s research. The effort combines analysis software with access to research data that may be unpublished, proprietary or governed by individual institutions. C-BRAIN says the tools are intended to help scientists generate and test hypotheses more quickly; the announcement describes research capabilities and goals, not a new treatment or clinical result.
Combining data without moving it all into one place
C-BRAIN launched in November 2025. Led by neurologist Randall Bateman, it brings together 18 member institutions across academic, philanthropic, pharmaceutical and research organisations. Its work addresses a practical challenge in biomedical research: relevant results and datasets can be scattered across organisations, while synthesising them takes specialist time.
Some data are deposited directly into secure computer systems. For data governed by member institutions, C-BRAIN also uses a federated approach: AI models can learn from data at each institution without that underlying data leaving its systems. The consortium describes its setup as a hybrid of centralised and federated computing, intended to let researchers work across data sources while respecting institutional governance and patient privacy.
C-BRAIN currently runs 64 NVIDIA data-centre GPUs on its premises alongside cloud-based nodes on Microsoft Azure. It says additional NVIDIA GPU capacity, including Blackwell systems, is planned as the consortium expands. C-BRAIN is also working with NVIDIA to incorporate the BioNeMo open developer platform and the MONAI framework into its research pipelines.
Two open tools target different research tasks
C-BRAIN researchers have built OpenScientist with Lawrence Berkeley National Laboratory and InsightEngine with 387 Labs. Both are described as open-source tools, but they address different parts of the research process.
OpenScientist is an agentic AI system that takes a scientific question and dataset, then semi-autonomously proposes hypotheses, runs analyses and produces outputs intended to be verifiable and auditable. InsightEngine is a research platform that tests hypotheses against scientific literature, analyses user-provided datasets and explores a biomedical knowledge graph. C-BRAIN says it checks claims against evidence to help researchers identify errors before committing to experiments.
NVIDIA’s account of published case studies says OpenScientist completed analyses in minutes that domain scientists estimated would take weeks to months, at a cost of less than $10 per run. C-BRAIN says its current reach exceeds 200 research groups and estimates that the capability could free more than one million hours of research capacity annually if deployed at scale. Those figures describe reported workflow results and a projection, rather than measured gains across all participating groups.
The consortium’s longer-term plan is to build a multimodal biological foundation model using neuroimaging, multiple types of molecular data, fluid biomarkers and longitudinal clinical measures. By its third year, it aims to identify causal mechanisms involved in brain ageing and Alzheimer’s disease. Those are research objectives; the tools’ reported current role is to analyse data and support hypothesis generation, steps that may contribute to discovery but are distinct from demonstrating a new prevention or treatment.