Q: How do you stay updated with the latest advancements in both the medical and computer science fields?
A: For advancements in medical science, I look up updates from UpToDate, a clinician reference with latest guidelines and treatment protocols; Medscape, an online medical news aggregator site; and The New England Journal of Medicine (NEJM). For computer science, I frequently visit sites like The Hacker News and TechCrunch.
Q: Can you share how your background in pharmacology and computer science has influenced your teaching methods at AUC?
A: Computational neuroscience has influenced my pedagogy. The brain has a natural tendency to seek patterns and build hierarchical mental models. I believe that teaching strategies should focus on explicitly connecting new information to existing knowledge and highlight organization. Starting with big-picture concepts before diving into details aligns with how our neural architecture processes information. I have been creating concept maps for students to help them with this. Since joining AUC, I have created more than 100 concept maps that connect with each other to form a network of information.
Q: How do you balance the use of traditional teaching methods with new technological approaches? What challenges have you faced while implementing new technologies in your teaching, and how have you overcome them?
A: Currently I’ve only been using AI and other technology to create resources for students. I do not directly expose the technology to students so that I can ensure I review all materials fully before students are exposed to the information. This ensures that students don’t get inaccurate information or get overloaded with information.
Q: Can you share an example of a new case scenario or multiple-choice questions that you’ve created using AI tools?
A: I’ve been using generative AI in creating mini-cases and multiple-choice questions. I believe I have now devised an efficient way of creating multiple-choice questions. I think of a possible multiple-choice question and instead of prompting the large language models (LLM) to generate the question, I only use the prompt to generate a clinical vignette. Once the vignette is created, I add logic and create distractors. This method increases productivity without sharing my intellectual property (the logic in the question) to be added to the training data of the LLM.
Q: What feedback have you received from students about your innovative teaching methods?
A: The student feedback so far has been phenomenal. I have been awarded the Professor of the Semester award for both semesters that I’ve been here!
Q: What role do you think student feedback should play in shaping the use of technology in education?
A: Student feedback is very important, especially with the introduction of new technologies. A tech company would collect a lot of data and feedback from end users to optimize the use of its technology. This user experience research, which comes under a branch of computer science called human-computer interaction, is vital. Similarly, I believe that a lot of pilot tests and data have to be collected and analyzed from students before fully integrating a new technology, especially generative AI. The initial optimism for new technology should also be weighed against established educational objectives and research-backed practices in pedagogy.
Q: You’ve expressed some reservations about AI in student-facing applications. Can you discuss your concerns in more detail?
A: Just like how a clinician would hesitate to use AI for direct patient care, as a teacher, I am cautious of using generative AI directly in a student-facing application. Most commercially available LLMs (ChatGPT, Gemini, Claude, etc.) are not specifically trained on clinical data and there could be factual errors in their responses. I am more worried about wrong reasonings they could provide students with, as LLMs are prone to hallucinations. AI hallucinations occur when language models confidently generate false or inaccurate information while attempting to respond to queries. Like a student making up an answer on a test when they’re unsure, AI systems can produce content that seems plausible and well-structured but is factually incorrect or entirely fabricated.
Q: How do you think your approach to using technology in education sets you apart from other educators in your field?
A: As a medical educator, I use technology as a means and not an end goal. I’m particularly mindful about avoiding using technology for technology’s sake. I choose technology based on its ability to improve conceptual understanding of the material, reduce student anxiety and stress and help improve learning efficacy.
Q: In what ways do you believe AI and other technologies can enhance faculty productivity?
A: I believe AI can help increase faculty productivity by aiding in new content creation (e.g., summary slides), creating practice quizzes and clinical cases, and possibly grading assignments in the near future.
Q: What are some of the ethical considerations you think are important when integrating AI into medical education?
A: AI uses training data from all over the internet and possibly some copyrighted material. It is important to review the information thoroughly to make sure there are no factual errors and the content provided to our students is not plagiarized.
Q: Could you share more about your involvement in curricular development and refinement at AUC?
A: Since I’m new to AUC, I am not directly involved with the curriculum committee yet. However, since I love to do data analysis with Python, I have been crunching numbers from past student data and assessments to identify weak areas and other factors that could be improved. Recently I analyzed more than 1,300 test items from the past three semesters to determine certain topics that require additional student support. This information was used to create a new pilot course, which is currently being evaluated. The amazing thing about this Python data analysis is that once I have the scripts ready, I can do minor alterations and analyze data for the whole school in seconds!
Q: Looking forward, what are your major goals for integrating technology into your teaching at AUC and which technologies are you most excited about?
A: These are some of the new tools I am excited about: Napkin is a tool that can automatically create visual diagrams from lines of text. Many students prefer visual information to reading long lines of text. There are possible AI integrations with Obsidian, a popular note-taking tool with backlinking support. I hope we could use this to create an AI bot that could automatically suggest similar topics or questions based on student responses or grades.
Q: What advice would you give to students interested in a career at the intersection of medicine and technology?
A: If you want to be a clinician with an interest in technology, a lot of research opportunities are out there in patient-facing technologies like augmented reality to improve patient outcomes, creation of mobile applications, data science in electronic health records, etc. I suggest you do some courses on sites like Stanford Online or Coursera on a field you are interested in to have some basic foundational knowledge to participate in such research activities. I would love to hear your ideas and discuss them over coffee.
Dr. Abialbon Paul does more than teach medicine — he’s transforming how it’s taught. By blending technology with thoughtful, student-focused strategies, he’s preparing AUC students to thrive in medical school and beyond. His work is a powerful reminder that the exceptional education at AUC goes beyond passing along knowledge — it’s about inspiring curiosity and empowering future physicians. If you’re looking for a place where you’ll be challenged, supported, and inspired, Dr. Paul and AUC are ready to welcome you.