Clinical Mind AI Research Presented at Stanford’s Big Ideas in Medicine Conference 2026
Max Ruiz, a medical student at Stanford School of Medicine and a research assistant with Clinical Mind AI, presented research on instructors’ perspectives on AI-simulated patients at the fourth annual Stanford Big Ideas in Medicine Conference. The event took place on September 11–12, 2026, at the Li Ka Shing Center for Learning and Knowledge.
His poster, “Beyond Access: AI-Simulated Patients and the Scalability of Clinical Reasoning Feedback,” presented findings from interviews with 51 medical instructors who used the Clinical Mind AI platform. Each instructor authored a clinical case, interviewed the resulting AI-simulated patient, and then participated in a semi-structured interview. The research examined which types of student performance analytics instructors valued, the barriers they anticipated when incorporating the platform into teaching, and how their feedback could inform future development.
A central finding concerned instructors’ interest in understanding how students approach a clinical encounter. Instructors valued analytics related to communication and questioning above other types of performance information. Their responses suggested that a useful contribution of the platform could be helping educators understand learners’ reasoning processes through the questions they ask and the way they conduct an interview. This finding highlights the potential value of feedback that examines how students work through a case, alongside assessment of the diagnosis they ultimately reach.
The interviews also identified practical challenges to adoption. Many instructors described the time required to author a clinical case as a major barrier, underscoring the importance of faculty workload when introducing new educational tools. Their feedback points to streamlining case creation as a potential priority for future platform development. Making repeated practice available to students also requires supporting the educators who design and integrate those experiences into their courses.
During the poster session, several attendees asked whether AI-simulated patients were intended to replace standardized patients. Ruiz emphasized their potential to complement existing training by providing additional opportunities for clinical reasoning practice. He explained that in-person encounters with standardized patients remain valuable, while Clinical Mind AI could allow students to practice more frequently than is typically feasible with live actors. The recurring question became an important part of his conversations with attendees about the platform’s educational purpose.
Most instructors interviewed envisioned using Clinical Mind AI in undergraduate medical education or early clinical training. They saw a role for the platform as students begin developing the skills needed for clinical encounters. Ruiz also shared his perspective as a medical student, expressing optimism that AI-simulated patients could offer a low-pressure environment for repeated practice. Such opportunities could help learners build a foundation in clinical reasoning before progressing to interviews with real patients.
Reflecting on the conference, Ruiz highlighted how much the framing of AI in education matters. Questions about replacing standardized patients encouraged him to lead with the platform’s role in complementing existing learning experiences. He was also struck by how many clinical instructors he met already had experience with AI teaching tools. Although their perspectives varied, he encountered broad interest in the potential of these technologies when used responsibly. These discussions connected the research findings with educators’ practical questions about feedback, faculty workload, and the place of AI-simulated patients in medical training.
