AAN 2026: How to Use AI Tools in Neurology Without Compromising Clinical Judgment

In an interview with Pharmacy Times at the 2026 American Academy of Neurology (AAN) Annual Meeting, James "Trey" Bateman, MD, MPH, associate professor of neurology at Virginia Commonwealth University (VCU) and medical director of the Alzheimer and Cognitive Disorders Clinic, identified the most immediate clinical applications of artificial intelligence (AI) in neurology as ambient listening and scribing tools, AI-enhanced clinical decision support within the electronic medical record, imaging software assistance, and natural language processing–driven clinical trial recruitment that can surface eligible patients from unstructured data.

Pharmacy Times: Can you introduce yourself and explain your current role?

James "Trey" Bateman, MD, MPH: Hi. My name is Trey Bateman. I'm a behavioral neurologist and associate professor of neurology at Virginia Commonwealth University School of Medicine. I'm also the medical director of the Alzheimer's and Cognitive Disorders Clinic.

Pharmacy Times: Where are you seeing AI make the most immediate and meaningful impact in neurological care right now?

Bateman: I think there are a couple of areas that really stand out. As a practicing clinician, I've seen more and more AI-driven tools being implemented within clinical workflows, and they are getting better and better. Many of us now have access to and have used AI scribes, or so-called ambient listening technology. They all vary somewhat—some seem to be better than others—but for many clinicians, they can meaningfully reduce workload, including so-called "pajama time" spent finishing dictations after hours. We're also seeing more capable clinical decision support and summarization functions within the medical record. We recently gained access to a new system at VCU that is quite impressive in its ability to draw on the context of a patient's past data and help clinicians reason through complex diagnoses. We're also seeing significant AI support in neuroimaging—both in imaging software and in supporting image interpretation. Within medical education, AI is enabling dynamic personalization of educational content, helping students focus their practice on areas where they are struggling. Students are already engaging AI agents directly—asking them to explain a difficult concept or present it from a different angle. Finally, one area I find particularly exciting is AI's potential to support clinical trial recruitment and engagement. It is extremely difficult to enroll enough patients in our trials, in part because so much relevant data [are] unstructured. AI using natural language processing can read that unstructured data and generate realistic lists of patients who may have expressed interest in research and appear to meet trial eligibility criteria.

Pharmacy Times: How is AI influencing clinical decision-making in neurology—and what should clinicians be cautious about as they begin to rely on these tools?

Bateman: AI, as many people are using it, can seem very impressive. The volume of data it processes and the confidence with which answers are sometimes delivered can feel almost overwhelming to those unfamiliar with the technology. I think clinicians need to recognize that, at the end of the day, they are the arbiter of the clinical decisions they are making. Right now, everyone I know is advocating for a person-in-the-loop or clinician-in-the-loop approach to decision-making. There are cases where an AI-driven algorithm may give you an answer that appears to make sense based on the data it has access to—but we must all remember: garbage in, garbage out. If the input data [are] unreliable, you may get a wrong answer, and expert clinicians still have a critical role in troubleshooting that. The important thing is to not allow AI to operate as a black box. Clinicians remain responsible for the decisions and diagnoses they make, and maintaining their own independent thought process—rather than deferring to AI to the exclusion of their own clinical reasoning—is still essential.

Pharmacy Times: Beyond clinical care, how is AI reshaping neurology research and medical education?

Bateman: Students are often at the vanguard of these changes, and we're seeing AI influence medical education both formally and informally. On the formal side, we are using AI for content development and curriculum management—for example, you could feed all of the lectures medical students receive over the course of a year into an AI system and ask how well the curriculum is addressing key topics like nutrition or other core domains. Without AI, that kind of audit would require poring over enormous amounts of information and finding some way to tally it manually. AI gives us a much better bird's-eye view of the content we're delivering. On the informal side, students have always sought out supplementary learning resources, and they are now using AI agents to summarize difficult chapters, generate podcasts from study material, or get alternative explanations of challenging concepts. This is, in some ways, the Wild West—we don't have full control over how students are going to use these tools. What is important is that we give them a clear understanding of AI governance at their institution, so they are not inadvertently feeding sensitive information into unsupported systems. Many institutions now have approved AI programs, and ensuring students are using those sanctioned platforms is increasingly important.

Pharmacy Times: What do neurologists and pharmacists need to understand about AI today to stay ahead of where the field is heading?

Bateman: First and foremost, neurologists and pharmacists need to recognize that AI is here, and it does not appear to be going anywhere. Learning to responsibly and appropriately use the tools that are available is important for everyone in the field—because if you don't, you will find yourself falling significantly behind. People are going to come to this from different starting points in their journey with AI. Those without a background in coding or the technical underpinnings of these systems don't need to learn all of that, but they do need to understand what AI is actually doing. Just as with any new technology, it is important for neurologists and pharmacists to stay abreast of what's possible and what's appropriate to use. Specialty societies—including the AAN, which has an AI governance group—and individual institutions are providing guidance on policies and appropriate use. Knowing what the rules are allows us to explore and adapt our own workflows within an appropriate framework. Ignoring AI entirely is simply not a viable option at this point.