Overview
- The organizations publicly announced the collaboration on Tuesday to develop a purpose-built, frontier AI model trained on Mayo Clinic’s de-identified, longitudinal clinical records.
- Mayo Clinic will retain ownership of the model and conduct initial real-world testing and clinician validation inside its clinical environment to track accuracy and safety.
- Microsoft will supply AI engineering and cloud infrastructure and intends to expose the validated model via Azure Foundry APIs so other health systems and developers can use it.
- Leaders say the effort aims to support earlier diagnosis, more personalized treatment and clinician tools, but they caution that training, validation and trust for high-stakes uses will take many years.
- The project builds on Mayo’s existing AI work and responds to concerns about general-purpose chatbots giving unreliable health advice, and it could change how patients and clinicians access evidence-based guidance if it proves reliable.