Johns Hopkins and OpenEvidence Collaborate on AI for Medical Education and Clinical Care

Johns Hopkins Medicine is partnering with OpenEvidence to develop artificial intelligence systems for clinical care and medical education. The collaboration will explore ways to combine medical evidence, individual patient histories and Johns Hopkins’ clinical best practices to support physicians’ decision-making and examine how medical training should adapt to the increasing use of AI.

Applying AI to Clinical Decision-Making

The organizations plan to develop what they call “human-first AI systems” that help clinicians retrieve relevant medical knowledge, interpret it in the context of individual patients and make clinical decisions. The specific capabilities of the systems are still being defined, according to Sara Murray, M.D., chief medical informatics and AI officer at Johns Hopkins Medicine.

Murray said the partners intend to identify clinical applications, develop solutions that fit into clinicians’ existing workflows and evaluate their effects on clinical decision-making, care quality and patient outcomes.

The collaboration also reflects an interest in moving beyond standalone medical information tools towards AI that is integrated into the delivery of care.

Preparing Physicians for AI-Assisted Medicine

Johns Hopkins Medicine and OpenEvidence believe the expanding use of AI will require changes to clinical practice and medical education. Although AI systems can retrieve and synthesise medical information rapidly, the organizations emphasize the continued importance of clinical reasoning, contextual judgment and the human aspects of patient care.

Sanjay Desai, M.D., vice dean for education at Johns Hopkins Medicine, said academic institutions have an opportunity and responsibility to study the consequences of new technologies and ensure that physicians remain adequately trained.

Desai also highlighted concerns about cognitive offloading, which involves relying on AI for competencies that physicians are expected to develop during training. He said educators need to distinguish between skills that can be offloaded to AI, those that can be augmented by the technology and those that remain inherently human.

He further stressed the importance of evaluating AI tools and monitoring their implementation to determine whether they meet expectations for clinical practice.

Retaining Essential Medical Skills

The partnership will focus on developing a medical education model that emphasizes pathophysiology, evidence-based medicine, clinical reasoning, diagnosis, and patient communication.

Desai said AI could make medical training more personalized by identifying gaps in students’ clinical experience and tailoring instruction to their individual needs. Potential applications include tracking students’ clinical exposure, providing simulations to address missing experiences, and identifying learners who may require additional support.

He also described the possibility of shifting medical education from a time-based structure towards competency-based training. Under this approach, assessments would focus on whether physicians had mastered the necessary skills rather than relying on the number of years spent in training.

Johns Hopkins Medicine comprises the Johns Hopkins University School of Medicine, established in 1893, and the Johns Hopkins Health System. Desai said the institution has an opportunity to help shape medical education as AI becomes more widely incorporated into training and clinical practice.

OpenEvidence’s Existing Clinical Platform

OpenEvidence operates a clinical decision-support platform and medical search engine for healthcare professionals. According to the company, 1.12 million licensed, verified U.S. clinicians, including physicians, nurses, nurse practitioners, and physician assistants, were using the platform as of September.

The company has also integrated its generative AI tools into several academic medical centres and health systems, including Mount Sinai Health System, Cedars-Sinai Health System, Penn Medicine, NewYork-Presbyterian and its affiliated medical schools, AdventHealth and Northwestern Medicine. At Northwestern Medicine, the platform is being embedded directly into its electronic health record system.

Research and Product Collaboration

OpenEvidence founder and CEO Daniel Nadler confirmed that the partnership includes research and development and product collaboration but did not disclose commercial arrangements.

Travis Zack, M.D., Ph.D., chief medical officer of OpenEvidence, said the collaboration would address how AI changes the role of physicians and how that transition should be shaped around patients.

The organizations have not yet specified the particular clinical tools they will develop together. Their work will encompass both the use of AI in care delivery and the preparation of future physicians for an environment in which the technology plays an increasing role.

Johns Hopkins and OpenEvidence: Advancing AI in Healthcare

Artificial intelligence is increasingly influencing how healthcare professionals learn, evaluate medical evidence, and make clinical decisions. The collaboration between Johns Hopkins and OpenEvidence reflects growing interest in using AI to support medical education and clinical care.

Johns Hopkins is widely recognized for its contributions to medical education, clinical research, and patient care. OpenEvidence focuses on AI-powered access to medical knowledge, helping healthcare professionals find relevant evidence and research to inform clinical questions.

By bringing medical expertise and AI technology together, the collaboration highlights an important direction for healthcare innovation: making trusted medical information more accessible while preserving the role of professional judgment.