A consortium led by Mayo Clinic, Imperial College London, and BioForge AI unveiled the Global Federated AI Marketplace (GFAM), a platform enabling researchers to train AI models on patient data without transferring raw records. GFAM uses secure enclaves, differential privacy, and standardized data interfaces (FHIR, DICOM) to ensure privacy and governance. The pilot program covers radiology, pathology, and clinical decision support across five hospitals in North America and Europe, with regulatory oversight from participating health systems.

Impact Analysis

Benefits include privacy-preserving access to diverse patient data, faster AI experimentation across institutions, and more generalizable models. It can reduce data silos and accelerate clinical AI deployment. Risks include governance and bias management, data interoperability challenges, and the need for robust security, consent, and regulatory alignment to scale responsibly.