The Art of Prompt Engineering: Optimizing Chatbots in Medical Education

Thomas Heston and Charya Khun

In a recent paper published in the peer-reviewed journal “International Medical Education,” Dr. Thomas F. Heston and medical student Charya Khun, MA, from the Elson S. Floyd College of Medicine explore the potential benefits and challenges associated with the utilization of chatbots like ChatGPT and Google Bard in medical training. While these AI systems possess vast training databases of great value, careful prompt engineering is required to properly use them for educational purposes.

Prompt engineering is the process of creating useful prompts that yield accurate and desired results. Thoughtfully crafted prompts offer contextual information, assign specific roles, and break down complex instructions. Paying meticulous attention to detail when crafting a prompt can mitigate the AI’s inclination at times to fabricate responses. While chatbots should never serve as the primary source of information, through prompt engineering they can prove invaluable in generating practice questions, investigating novel concepts, and simulating clinical scenarios.

This paper directly addresses the significant, growing concerns pertaining to the ethical use of AI chatbots. Guardrails must keep pace with the growth of these AI engines. Notwithstanding these challenges, well-optimized prompts can aid educators in carefully harnessing the capabilities of these potent tools in an ethically responsible manner.

Heston TF, Khun C. Prompt Engineering in Medical Education. International Medical Education. 2023; 2(3):198-205.