History
Clinical Mind AI began in 2023 at Stanford University as a student-led project. The first prototype was developed by students Marcos Rojas (MD, PhD candidate in education) and Henrik Marklund (PhD candidate in engineering) as a class assignment for CS372: Artificial General Intelligence for Reasoning, Planning, and Decision Making, taught by Professor Edward Y. Chang. This early version used generative AI and a conversational chatbot to simulate a patient encounter, allowing learners to converse with virtual patients and begin practicing clinical reasoning, a foundational skill for healthcare practitioners.
Having obtained his MD in Chile and now researching medical education in the US, Dr. Rojas deeply understood the diverse needs of medical educators worldwide. He knew from experience that few medical schools globally can afford to provide the type of clinical reasoning practice, despite this being the most authentic training method. The proof-of-concept developed by the team provided compelling evidence that large language models (LLMs) could simulate patients accurately and effectively—at a fraction of the cost of human actors.
Following this initial prototype, the project received early recognition and support through student innovation opportunities, including the Stanford Learning Design Challenge. This support enabled further development of the platform. Chinat Yu, an education master’s student, later joined the team and expanded Clinical Mind AI's capabilities, allowing instructors to generate AI-simulated patients based on their own clinical cases.
Clinical Mind AI continued to develop through a collaboration between the IDEAL Lab and the Chariot Lab. The IDEAL Learning Lab, led by professor Shima Salehi, focuses on designing technology-rich learning environments that support learners from all backgrounds, with particular emphasis on teaching authentic problem-solving and empowering instructors with AI tools to create equitable, adaptive, and affordable instructional materials. The Chariot Lab, led by Dr. Thomas Caruso, harnesses immersive technology to drive the future of medical education, enabling healthcare professionals to learn in highly realistic scenarios that refine their skills and deepen their understanding.
These complementary research perspectives—one rooted in the learning sciences and technology-mediated instructional equity, the other in clinical simulation and medical training—helped shape Clinical Mind AI as a comprehensive AI-powered learning environment to: 1) empower instructors to create customized clinical cases tailored to their students' diverse contextual needs; 2) simulate virtual patients based on these cases for students to interact with as an authentic and realistic opportunity to practice clinical reasoning; and 3) provide instructors and medical students with real-time, targeted feedback that supports the development of clinical reasoning required for expert, high-quality patient care.
Between 2023 and 2026, the platform was piloted with 20 academic institutions worldwide. During this period, it also became an active engine for scholarly inquiry, with faculty, research staff, and students contributing to a growing body of peer-reviewed work. Research contributions span presentations at leading conferences and publications in academic journals, with work focusing on the potential of AI to enhance clinical reasoning, improve teaching methodologies, and shape the future of medical education. Published studies examined topics including differences in clinical decision-making between medical students and expert clinicians, the role of AI in medical licensing examinations, and a systematic review of artificial intelligence in medical education.
The platform was supported by Human Artificial Intelligence and Stanford Impact Labs awards in 2024, in addition to other innovation programs. It was presented at venues including the Ottawa Conference, the AI+Education Summit at Stanford, and received the Reimagine Education Gold Award in 2025.
In 2026, Stanford identified Clinical Mind AI as an institutional resource, affirming its role in advancing education research, effective and equitable teaching, and global impact.