AI Is Coming to Academic Medicine — Are You Ready?
This Is Not a Distant Future
If you work in academic medicine, artificial intelligence is no longer a topic for futurists and keynote speakers. It is here. It is in your radiology reads, your EHR inbox, your students' study habits, and increasingly in the grant proposals landing on review committees.
The question is not whether AI will change your work. It already is. The question is whether you are paying enough attention to shape how it changes things, or whether you are going to wake up one morning and wonder what happened.
The Three Fronts
AI is hitting academic medicine on three fronts simultaneously, which makes it uniquely disruptive for faculty who already juggle clinical, research, and teaching roles.
Clinical care. Large language models and decision-support tools are being embedded in workflows across specialties. A 2023 study in Nature Medicine showed that an AI model matched or exceeded physician performance on clinical reasoning tasks. That does not mean it is replacing you. It means the nature of your clinical expertise is shifting. Pattern recognition is becoming less valuable. Judgment, communication, and the ability to work alongside these tools is becoming more valuable.
Research. AI is accelerating literature review, data analysis, and even hypothesis generation. If you are a researcher who has not explored how these tools can augment your workflow, you are leaving productivity on the table. If you are a PI managing a lab, your trainees are already using them whether you have talked about it or not.
Education. This may be where the disruption is most immediate. Students are using AI to draft papers, study for boards, and generate clinical summaries. The old models of assessment are breaking down. Faculty who teach need to rethink not just how they evaluate learning, but what they are actually teaching students to do.
The Leadership Vacuum
Here is what concerns me most. At many institutions, the people making decisions about AI adoption are not the faculty who understand the clinical and educational nuances. They are administrators and IT teams working from a technology-first mindset. That is not a criticism of those teams. It is an observation that physician voices are needed in these conversations and are often absent.
Eric Topol has been writing about this for years. In Deep Medicine, he argued that AI's greatest potential in healthcare is not replacing physicians but freeing them to do the deeply human work that machines cannot. I think he is right. But that future does not happen automatically. It happens because physicians step into leadership roles and insist on it.
If you are a faculty member with any influence over how your department or institution adopts AI, you have an obligation to be in the room.
What You Can Do Right Now
You do not need to become a data scientist. But you do need to become literate. Here are a few starting points.
Use the tools yourself. Spend an afternoon with a large language model. Ask it clinical questions in your specialty. See where it is impressive and where it falls apart. You cannot evaluate something you have never touched.
Talk to your trainees. They are already using AI. Find out how. You might learn something, and you will definitely surface questions your program needs to address.
Join the conversation at your institution. If there is a committee or task force on AI, volunteer. If there is not one, ask why. The decisions being made now will shape the next decade of academic medicine at your institution.
Read broadly. Topol's Deep Medicine is a good starting point. Follow what organizations like the AMA and AAMC are publishing on AI policy. Stay curious rather than anxious.
This Is a Leadership Moment
Every major shift in medicine creates a window where the people who engage early have outsized influence. AI in academic medicine is that window right now. The faculty who lean in, who learn enough to ask good questions, who advocate for thoughtful implementation, those are the ones who will shape what comes next.
The ones who wait will adapt eventually. But they will be adapting to a system that was built without their input.
If you are thinking about how AI fits into your career trajectory or how to lead your department through this shift, that is exactly the kind of challenge I help academic physicians work through. I would welcome the conversation.
