Travis Zack, MD, PhD, who grew up in Lahaina, Hawaii—a town that suffered extensive wildfires in 2023—has taken a remarkable and unconventional route to a career in medicine. His journey is influenced by personal experiences, a strong family background, and mentors who have guided him along the way. Recently, he has developed a keen interest in how artificial intelligence (AI) can assist clinicians in managing the increasingly intricate world of medical knowledge.
On family legacy: My parents instilled the importance of education in me from a young age.
On mentors: An undergraduate mentor encouraged me to explore biophysics instead of focusing solely on physics, which significantly altered my career path.
On a career in research and clinic: While working in computational genomics at the Broad Institute, I was inspired by MD-PhDs and physicians who were making meaningful contributions to research while also caring for patients. It was then that I realized I wanted to pursue both avenues.
On the role of AI in health care: The standard for AI in healthcare should not just be convenience; it must also prioritize transparency and verifiability.
Dr. Zack serves as an Assistant Adjunct Professor at UCSF and the Chief Medical Officer of OpenEvidence, a medical information AI platform widely utilized by healthcare professionals across the United States. His expertise encompasses computational biology, oncology, machine learning, and clinical decision support.
In this segment of The ASCO Post’s Living a Full Life series, Guest Editor Jame Abraham, MD, FACP, engaged Dr. Zack in a discussion about his Hawaiian upbringing, influential mentors, his shift from academic oncology to health AI, and his vision of generative AI’s impact on medical education and oncology. Below are excerpts from their conversation, edited for brevity and clarity.
Tracing the Early Beginnings
Jame Abraham, MD, FACP: To begin, could you share your story? You were born in Hawaii. How did your upbringing influence your journey into medicine?
Travis Zack, MD, PhD: I grew up in Lahaina, a small town on the northwest coast of Maui. My mother was a preschool teacher, and both of my parents were actively engaged in the community. Although they weren’t academics, they strongly valued education and instilled that belief in me from an early age.
During my college years at UC Berkeley, I was unsure of my career path. I started as a psychology major, switched to philosophy, and eventually found my niche in physics. While it was the toughest subject I tackled, it ignited my passion for research and understanding fundamental principles—decomposing complex problems, challenging existing assumptions, and seeking more efficient solutions to health system challenges.
Jame Abraham, MD, FACP
Dr. Abraham is the Chairman of the Department of Hematology and Medical Oncology at the Cleveland Clinic and a Professor of Medicine at the Lerner College of Medicine.
One of my undergraduate mentors recognized my background and suggested that I explore biophysics rather than traditional physics. That insight changed my course significantly. This mentor, John Clarke, PhD, ScD, was awarded the Nobel Prize in Physics in 2025 for his groundbreaking work in quantum physics.
Initially, I was keen on imaging techniques at the single-molecule or atom level. Coming from a small town with no major hospitals or research facilities, I was unaware that physicians could also dedicate themselves to research work. At first, I saw research and medicine as two distinct paths.
Discovering Oncology
Dr. Abraham: What led you to change that perspective?
Dr. Zack: During my graduate studies in biophysics, my mother was diagnosed with breast cancer. While engaged in somewhat abstract imaging research, I began questioning the real-world impact of my work.
This experience motivated me to pursue rotations in cancer research. At the Broad Institute, I worked in computational genomics, surrounded by MD-PhDs and physicians who combined meaningful research with patient care. It was then that I realized I wanted to embrace both aspects. After earning my PhD in Biophysics from Harvard, I attended medical school through the Health Sciences and Technology program at MIT and Harvard Medical School. I completed my residency and fellowship in oncology at UCSF, specifically at the UCSF Helen Diller Family Comprehensive Cancer Center.
A Family Experience With Cancer
Dr. Abraham: Your mother’s battle with cancer seems to have significantly impacted your journey.
Dr. Zack: Absolutely. While my mother is doing well now, her experience was pivotal in defining my path. It was later discovered that our family has a BRCA2 mutation, which we only realized after her second brush with breast cancer. My grandfather had also died young, thought to have a gastrointestinal malignancy, suggesting he too may have been a BRCA2 carrier. This familial context makes my role in oncology both professional and deeply personal.
Mentors Along the Way
Dr. Abraham: Who were some of the influential mentors you encountered along your journey?
Dr. Zack: I benefited from numerous mentors. During my graduate studies, many recognized my quantitative strengths while also guiding me to strengthen my biological knowledge for impactful translational work.
One critical figure in my medical journey was Eliezer Van Allen, MD, at Dana-Farber, who introduced me to machine learning. Later, I had the fortunate opportunity to work with Atul Butte, MD, PhD, at UCSF. He not only nurtured my interest in machine learning and real-world data but also emphasized the importance of developing research tools that could enhance patient care. Tragically, Atul passed away in June 2025 after a two-year battle with cancer at just 55 years old. His contributions to big data in science and healthcare will be felt for generations, and his mentorship was invaluable.
From Academic Oncology to OpenEvidence
Dr. Abraham: You were on a traditional academic trajectory. Can you tell us about OpenEvidence and what prompted your full-time transition there?
Dr. Zack: For a considerable time, I envisioned myself as a tenured professor. While at UCSF, I was enthusiastic about that traditional path. However, my collaboration with OpenEvidence revealed that the company was not merely publishing papers or engaging in technical work; it was actively reshaping medical practices. That was incredibly appealing. To date, we have facilitated over 100 million AI-powered clinical consultations for U.S. physicians and clinicians at the point of care.
I still engage in activities that are important to me, such as mentoring and focusing on research quality, but now I spend more time connecting with health systems, professional societies, and publishers. I continue to contribute to building a better healthcare system, but via a different approach and on a larger scale.
How Generative AI Is Being Used in Medicine
Dr. Abraham: What are your thoughts on generative AI and its current transformation of healthcare?
Dr. Zack: One of the primary functions of generative AI in medicine is to help clinicians sift through vast amounts of information, whether from medical literature or electronic medical records.
Humans are not inherently equipped to perform these extensive tasks efficiently. A community oncologist managing multiple cancer types can’t realistically retain all the published data on those diseases, especially as new information emerges at an incredible pace. Similarly, electronic patient records are becoming increasingly lengthy and fragmented, complicating the identification of vital information needed for each patient visit.
Generative AI can help overcome these hurdles through effective retrieval and synthesis of information. While these tools are not infallible, neither are humans. AI addresses a long-standing dilemma in medicine: the overwhelming amount of knowledge surpasses what any individual can access efficiently in clinical practice.
What OpenEvidence Was Built to Do
Dr. Abraham: Can you elaborate on what OpenEvidence aims to achieve and what excites you about it?
Dr. Zack: OpenEvidence is an AI platform designed to tackle the issue of precision medical information retrieval. Given the sheer volume of publications and the continuous flow of new research weekly, our goal is to help clinicians quickly find relevant information in response to their inquiries.
Simply generating answers isn’t sufficient; those answers must be rooted in appropriate, verifiable references. This entails identifying credible sources before generating responses. We also strongly believe that supporting and recognizing the contributions of individuals and organizations in medical knowledge creation is vital to maintaining the integrity of evidence-based medicine.
Hence, licensing and attribution are crucial. The objective is for clinicians to ask questions and receive timely, evidence-based answers accompanied by proper references. Increasingly, these responses also include original figures, pathways, and guideline flowcharts to clarify the support behind each recommendation.
Trust, Verification, and Clinical Use
Dr. Abraham: Trust is fundamental in medicine. How do you regard trust in AI?
Dr. Zack: Trust is crucial. Medical professionals make critical decisions based on the data they receive. Therefore, the standard must encompass more than fluency—it should emphasize transparency and verifiability.
This is why it’s essential to not just provide citations but also to make the evidential foundations visible whenever feasible. If a system references a guideline, figure, or clinical pathway, clinicians should have access to that source to review and comprehend how the conclusion was reached.
Medical Education in the AI Era
Dr. Abraham: Many trainees ponder whether AI tools will diminish the importance of memorization or even the relevance of physicians themselves. What is your take on this?
Dr. Zack: The role of memorization is indeed evolving. Physicians may no longer need to memorize every possible therapy option for every disease stage, especially given the rapid evolution of medical knowledge.
What will become increasingly critical is clinical reasoning grounded in a solid understanding of pathophysiology. Even as treatment protocols and side effects evolve, grasping the underlying biological principles is essential for determining the best patient-centric approaches.
In this way, AI tools underscore the human aspects of medicine. Physicians still need to interpret evidence tailored to the individual patient’s context—something AI cannot glean from trial results alone. Clinical judgement, context, and human insight will remain indispensable.
Looking Ahead for Oncology Practice
Dr. Abraham: Projecting 5 to 10 years ahead, how do you foresee AI reshaping oncology?
Dr. Zack: It’s already apparent how challenging it is for community oncologists to keep pace with the growing volume of specialized recommendations across various diseases. This challenge persists even at large medical centers, and it’s even more pronounced in community settings.
What excites me is the prospect that oncologists won’t need to invest excessive time poring over extensive documents or deciphering unfamiliar diseases before seeing patients. They will be able to turn to AI tools that quickly surface pertinent evidence, enabling them to apply their clinical expertise and judgement effectively.
I believe AI will also propel us toward precision therapy while supporting rather than replacing clinicians. This presents the opportunity to alleviate cognitive load, allowing physicians to focus more on patient engagement and less on information retrieval.
Words of Advice for the Oncology Community
Dr. Abraham: What final thoughts can you offer to the broader oncology community as the AI revolution unfolds?
Dr. Zack: It’s increasingly important for healthcare professionals to possess a degree of AI literacy, regardless of whether they directly utilize these tools. Patients are already engaging with AI technology—this is like an evolution following the era of “Dr. Google.” Oncologists will need to aid patients in interpreting the information they find and help differentiate the reliable from the unreliable.
Clinicians should engage with AI judiciously—neither blindly trusting it nor dismissing it entirely. A considered approach is essential; understanding what these systems excel at, where they might falter, and how they can enhance high-quality care is paramount.
When used responsibly, AI can mitigate information overload while safeguarding the core elements of oncology: clinician judgement, patient context, and the essential human relationships that underpin care.
Disclosure: Dr. Zack is the Chief Medical Officer of OpenEvidence.
