The “Health Informatics and Decision Support” lecture introduces informatics as the interdisciplinary study of how data are captured, processed, and applied to improve healthcare delivery. It outlines key informatics domains and their role in enabling modern systems, including electronic health records (EHRs), health IT infrastructure, interoperability standards, clinical decision support tools, and usability principles that shape how clinicians interact with technology.  

The “Artificial Intelligence in Health Care” session is a refresher lecture for senior medical students on how large language models process vast datasets to generate context-aware outputs that support clinical tasks. It highlights current and emerging applications such as documentation, diagnostic support, and patient communication, while emphasizing both benefits and risks, and the existing regulatory landscape for regulating the use of AI within and outside of health care. 

From their first weeks in medical school, students engage with MedSimAI, an AI-powered simulation platform that deliberate practice, self-regulated learning, and automated assessment through interactive patient encounters. Students learn to take complete histories with AI simulated patients with different presenting symptoms, and receive immediate, structure feedback using established frameworks such as the Master Interview Rating Scale.  

The “Illness Scripts” lecture in the Health, Illness, and Disease course emphasizes how clinicians use structured mental frameworks to organize knowledge and support diagnostic reasoning. Students practice generating prioritized differential diagnoses by integrating key features such as epidemiology, pathophysiology, and clinical presentation, while applying this approach to case examples to strengthen diagnostic accuracy. This lecture highlights how illness scripts can be strengthened by incorporating AI-supported insights into clinical encounters, enabling learners to synthesize diverse data sources such as patient history, epidemiologic patterns, and AI-generated suggestions. In practicing case-based diagnostic reasoning, students learn to integrate these inputs into coherent, patient-centered care plans while maintaining focus on the individual patient context. The session emphasizes that AI can broaden diagnostic consideration sets and reduce oversight, but effective use requires aligning algorithmic outputs with patient values, preferences, and clinical nuance.  

In the clerkships, students use skills developed in an AI-powered simulated environment to care for real patients and extend these skills in clinical settings by using AI-powered decision tools to integrate clinical guidelines, evidence summaries, and patient-specific data into more informed care plans in a variety of specialties. 

The “AI in Clinical Ethics Analysis” session explores the ethical dilemmas of integrating generative AI into patient care by uniquely asking students to pose ethical questions to a live, AI-powered chatbot trained with a persona of an ethicist. This occurs under the facilitation and guidance of an experienced human clinical ethicist, who challenges students to reflect on their conversations and interactions with the chatbot. 

An emphasis made in the “Health Informatics and Decision Support” lecture is the fundamental theorem of informatics, emphasizing that technology alone does not improve care, but rather the combination of humans and well-designed information systems working together enhances clinical decision-making, efficiency, and patient outcomes. 

Students integrate the use of AI-powered decision support tools during in-person teaching rounds with supervising faculty, who engage them in real-time “human-powered” clinical reasoning. This blended approach teaches students to thoughtfully incorporate AI insights while maintaining focus on the individual patient’s context, values, and needs, reinforcing both effective clinical reasoning and responsible technology use. 

A central theme of the “Illness Scripts” lecture is the balance between leveraging AI tools and preserving clinician responsibility. While AI can assist with generating differential diagnoses and identifying patterns, the lecture emphasizes that human clinicians must critically interpret these outputs, avoid overreliance, and remain accountable for final decisions. Through discussion of cognitive biases and diagnostic errors, learners see how AI may help reduce certain biases, yet also appreciate that clinical judgment, reflective thinking, and ethical responsibility remain essential to safe and effective patient care. 

The “AI in Clinical Ethics Analysis” session is designed as an interactive exchange between a clinical ethicist and an AI-based chatbot capable of participating in ethical reasoning. Within this session, students are encouraged to critically examine the ethical implications of AI in healthcare, including threats to privacy, challenges to meaningful informed consent, ambiguous lines of accountability, and unresolved questions around data ownership. They also engage in applying established ethical frameworks to assess emerging AI applications in clinical settings, with particular emphasis on critically appraising how generative AI may influence patient trust, clinical decision-making, privacy protections, and health equity.

In the  “Health Informatics and Decision Support” session, students learn how AI-enabled systems exist within the broader informatics ecosystem of EHRs, clinical decision support, and health IT infrastructure, and how these tools influence clinical workflows, decision-making, and ultimately patient outcomes and safety. The lecture also prepares students to use AI-driven insights such as predictive analytics and decision support outputs by demonstrating how structured data and interoperable systems enable meaningful analysis of patient care patterns. 

During clerkships, medical students interact daily with the electronic health record and its built-in alerts function, thereby using it as a real-time safety and quality support tools in the care of their patients. These include patient safety features such as such as drug–drug interaction alerts, allergy warnings, best practice advisories, and health maintenance reminders, which prompt students and clinicians to avoid errors. 

The “AI in Clinical Ethics Analysis” session is an interactive session shared between a practicing clinical ethicist and an AI-powered chatbot trained to engage in ethical discourse. As a part of this session, students identify key ethical challenges in AI use, including risks to privacy, limitations of informed consent, unclear accountability, and concerns about data ownership. Students also to apply established ethical frameworks to evaluate emerging AI technologies in clinical care, and critically assess how generative AI tools may impact patient trust, decision-making, privacy, and equity.