Microsoft and Mayo Clinic Build AI Model Trained Specifically for Medicine
Microsoft and Mayo Clinic are partnering to develop an AI model trained on medical records, clinical research, and the expertise of Mayo's physicians. The tool aims to help both patients and healthcare providers make better decisions about diagnosis and treatment.
Tens of millions of people now turn to AI chatbots for health advice. Most mainstream chatbots, however, are trained on broad internet content-meaning the medical guidance they provide can be inaccurate or even harmful. Mayo Clinic has previously warned that health information from these general-purpose chatbots is unreliable.
Mayo Clinic will own the resulting model. The organizations plan to first deploy it for clinicians at Mayo's hospitals to test for accuracy, then build a patient-facing AI assistant accessible through the hospital's online portal. They may eventually license the technology to other healthcare institutions.
Microsoft AI CEO Mustafa Suleyman said the project will take "many years" to train and refine the model to the point where it can be trusted for high-stakes health questions. Both organizations are making "very material, long-term commitments," though they declined to disclose spending.
Mayo Clinic's advantage lies in its data. Advanced AI models require large amounts of unique, high-quality information to improve their outputs. Mayo has anonymized patient records available for training and has already built smaller AI models to detect heart disease and diagnose pancreatic cancer.
Google, Anthropic, and OpenAI are also racing to dominate health advice. Google has rolled out an "AI Health Coach" for fitness and medical data review. But Mayo's decades of clinical experience treating complex conditions could differentiate this partnership.
Mayo Clinic CEO Gianrico Farrugia said the model could eventually help prevent disease, though he acknowledged "valid concerns" about AI in healthcare. "There's still so much need for better healthcare," he said. "We should embrace AI because it helps us get better results."
For healthcare professionals, this partnership signals where the industry is headed: AI for Healthcare is moving from general-purpose tools to models trained on specialized medical data. Understanding how Generative AI and LLM technologies work in clinical settings will likely become essential knowledge in healthcare roles.
Your membership also unlocks:
