50 Leading Global Thought Leaders on AI in Healthcare

Introduction

The most consequential debate in global medicine right now is not about a drug, a surgical technique, or a policy reform. It is about whether artificial intelligence will fulfil its extraordinary promise of transforming healthcare, or whether it will replicate and scale the inequities, oversights, and blind spots already embedded in the systems it is meant to improve. Both outcomes are possible. Which one prevails depends almost entirely on the quality of thinking that guides how AI is built, deployed, governed, and challenged.

Healthcare AI is no longer theoretical. According to research published in the journal Healthcare, approximately 80% of healthcare organisations now use AI applications in some form, and the global healthcare AI market is projected to expand from USD 26.6 billion in 2024 to USD 187.7 billion by 2030. Ambient documentation alone generated USD 600 million in 2025, more revenue and clinical adoption than any other category of clinical AI application. Meanwhile, the United States Food and Drug Administration has now cleared nearly 1,000 AI-enabled medical devices, covering everything from stroke detection on head CT scans to arrhythmia detection from single-lead electrocardiograms. The speed of deployment has outpaced the frameworks designed to govern it.

This list exists because navigating that gap requires following the people who are thinking most clearly about it. The 50 thinkers profiled here come from radiology and informatics, from machine learning labs and clinical wards, from Africa and Australia and Israel and Ireland, from patient advocacy and public health policy, from oncology and emergency medicine and the study of algorithmic bias. What they share is not a single perspective on AI. They disagree, often sharply, on questions of pace, regulation, equity, and clinical readiness. That disagreement is precisely why following multiple voices in this space matters. No single thinker has the full picture. The field is moving too fast and touching too many parts of medicine for that to be true.

Jonno White is a Certified Working Genius Facilitator and bestselling author of Step Up or Step Out (10,000+ copies sold globally). He works with healthcare leadership teams to build the communication, accountability, and alignment frameworks that allow the strategic decisions these thinkers advocate for to actually translate into results at ward level, board level, and everywhere in between. To bring Jonno in to work with your healthcare leadership team, email jonno@consultclarity.org.

Why AI in Healthcare Matters

The stakes of getting AI right in healthcare are different from the stakes in any other industry. A biased algorithm in retail recommends the wrong product. A biased algorithm in healthcare withholds care from patients who need it most. The landmark 2019 study by Ziad Obermeyer and colleagues, published in Science, documented how a widely deployed commercial algorithm used to identify high-need patients was dramatically undercounting the proportion of Black patients who qualified for care management, because it used historical healthcare cost as a proxy for clinical need, and Black patients at equivalent clinical severity generated lower costs than white patients due to structural barriers to access. The algorithm was encoding discrimination at scale, affecting millions of people, without any individual deliberately choosing that outcome.

This is why the conversation about AI in healthcare cannot be separated from the conversation about equity, governance, transparency, and the structures of trust that make clinical care possible. The people on this list are not simply enthusiasts for a technology that is genuinely impressive. Many of them are the most important critics of how that technology is currently being deployed. Several have testified before legislatures, founded non-profits to fill governance gaps, or spent their careers building the methodological foundations that others then draw on. Following them gives you a more accurate, more honest, and ultimately more useful picture of what AI can and cannot do in healthcare.

For more on navigating leadership challenges in healthcare organisations, explore my blog post '35 Essential Thought Leaders in Hospital Leadership Globally (2026)' at https://www.consultclarity.org/post/thought-leaders-hospital-leadership-global.

If your healthcare leadership team is ready to act on the ideas these thinkers champion, Jonno White works with health system leaders to build the team alignment and communication culture that makes complex change possible. Email jonno@consultclarity.org.

How This List Was Compiled

This list prioritises genuine contribution over institutional prestige. The people included here have produced original work, published in peer-reviewed venues, led organisations with real-world impact, or built movements that have shifted how the field thinks about a problem. Geographic diversity was a deliberate criterion: no single country represents more than 35% of the list, and voices from Africa, the Asia-Pacific, Europe, Israel, and the Global South are represented alongside the North American and UK voices who dominate most comparable lists. The selection process specifically sought out mid-career researchers, clinicians, and practitioners who are actively publishing and posting in 2025 and 2026, rather than recycling the same household names that appear on every other list. The result is a directory of people the reader may not yet have encountered alongside those who have shaped the field for a decade or more.

Category 1: The Governance and Standards Builders

These six people are doing the unglamorous work of building the frameworks, standards, and accountability structures without which responsible AI deployment in healthcare is impossible. They understand that the question is not whether AI should be used in healthcare, but how to govern its use in ways that protect patients, support clinicians, and hold vendors and health systems accountable.

1. Brian Anderson MD

Coalition for Health AI (CHAI)

The CEO and co-founder of the Coalition for Health AI, Brian Anderson is the architect of one of the most important governance experiments in global health AI. A Harvard Medical School-trained family physician with a background at MITRE Corporation, Anderson built CHAI from the ground up in 2021 to create consensus-driven guidelines and independent quality assurance frameworks for AI in health, filling a gap that both regulators and health systems acknowledged existed but could not fill alone. He brings a rare combination of clinical credibility, government experience from leading MITRE's work alongside the White House COVID Task Force, and a practitioner's understanding of what responsible AI deployment actually requires at the ward level.

His most cited institutional contribution is the CHAI Guidance on Responsible Use of AI in Healthcare, released in partnership with The Joint Commission in September 2025. This document, the product of thousands of hours of multi-stakeholder work, provides the first practical governance framework that health systems of any size can use to operationalise AI oversight. Anderson has been direct in congressional testimony and public forums that ignorance of algorithmic bias is not a defence, and that health systems have a duty of care to understand and audit the tools they deploy on patients.

2. Sherri Douville

TTIC (Trustworthy Technology and Innovation Consortium) / Medigram

The Chair of the Trustworthy Technology and Innovation Consortium (TTIC) and CEO of Medigram, Sherri Douville occupies a unique position at the intersection of AI standards, board governance, and healthcare accountability. As Chair of TTIC, she focuses on translating published AI governance frameworks into the board-ready artefacts and operational checklists that health system CIOs and chief executives can actually deploy in institutional practice. Her recent work centres on operationalising ANSI/HSI 2800:2025, the first American National Standard for AI governance in healthcare operations, published in December 2025.

Douville is the Series Editor for the Trustworthy Technology and Innovation book series published by Routledge/Taylor & Francis, one of the most substantial compilations of practical guidance on healthcare IT governance in print. She posts prolifically on LinkedIn about the gap between governance principles and governance practice, consistently directing her audience toward the structural and fiduciary accountability questions that boards and executives need to answer before deploying AI systems at scale. Her framing is distinctive: she argues that the cost of ungoverned AI deployment is not theoretical but financially quantifiable, citing documented settlements and federal court disclosure orders as evidence that the accountability era for health AI has already arrived.

3. Hamed Abbaszadegan MD

Stanson Health, Premier

A physician executive at Stanson Health and Chair of the HSI National AI Standards Committee, Hamed Abbaszadegan chaired the working group that produced ANSI/HSI 2800:2025, the landmark national standard for AI governance in healthcare operations. This work required building consensus across health systems, technology vendors, regulators, and clinicians with genuinely different interests and timelines. The resulting standard addresses board-level oversight, CEO accountability, and the practical steps health systems need to take to operationalise governance across their AI use cases.

Abbaszadegan's contribution is in turning the abstract principles of trustworthy AI into something an institution can actually audit. His clinical background in physician leadership and his operational experience across healthcare institutions gives him the dual perspective needed to build standards that work in practice rather than just in theory. He has been a visible speaker at HIMSS 2026 and the AI Standards Hub Global Summit, where he has argued that the 2026 moment marks the shift from AI ambition to AI accountability.

4. Brian Weiss

AWS Healthcare

At Amazon Web Services Healthcare, Brian Weiss leads partnerships and go-to-market work at the intersection of cloud infrastructure and clinical AI. His contribution to governance is less formal than Anderson's or Abbaszadegan's but no less important: he works at the moment when health systems move from theorising about AI to deploying it, and his public writing focuses on the gap between what health systems expect from AI and what the technology can currently deliver reliably and safely in regulated clinical environments.

Weiss has written specifically about the governance structures that organisations need to build before deployment, not after, including the human oversight layers, audit trails, and rollback protocols that distinguish responsible AI adoption from reckless implementation. His perspective from the cloud infrastructure layer gives him visibility into the reality of AI deployment across hundreds of health systems simultaneously.

5. Kameron Matthews MD

Cityblock Health

The Chief Health Officer of Cityblock Health, a value-based care provider serving Medicaid and Medicare beneficiaries, Kameron Matthews is one of the most important voices on the governance question that gets asked least often: who is responsible for AI outcomes when those outcomes affect the patients least able to advocate for themselves? As a co-chair of the Coalition for Health AI and a physician with a law degree, Matthews bridges clinical, legal, and policy dimensions of healthcare AI governance in ways that most contributors to this conversation cannot.

Her LinkedIn presence consistently raises questions about how AI standards need to be designed with equity at their centre, not added as an afterthought. She has argued that governance frameworks built primarily by and for large health systems will systematically underserve the populations who depend on safety-net providers, Medicaid plans, and community health centres, precisely because those organisations lack the resources and technical staff to operationalise frameworks designed for enterprise health systems.

6. Jonah Feldman MD

NYU Langone Health

A physician informaticist at NYU Langone Health, Jonah Feldman is one of the most thoughtful practitioners writing about AI governance at the level where it actually matters: inside a working health system, with real EHR infrastructure, real compliance requirements, and real clinicians who have neither the time nor the training to serve as AI auditors. His LinkedIn posts, which consistently attract substantive engagement from health IT professionals, cover the specific federal and state legislative landscape, tracing bills moving through Congress that would define AI practitioner status, create Medicare payment pathways for AI-enabled devices, and fund research into AI's impact on administrative burden.

Feldman published work in 2025 tracking how AI in healthcare is moving faster than regulation can keep up, and his analysis focuses on the institutional risk that health systems face when they deploy AI tools in a regulatory vacuum. His dual identity as a practising physician and a health IT leader gives him a perspective that is rare and valuable.

Category 2: The Machine Learning Researchers Building Fairer Systems

These nine people are doing the foundational research work that the rest of the field depends on: building algorithms, exposing their failures, developing new methods to make them more robust, and publishing the results where others can learn from and build on them.

7. Marzyeh Ghassemi

MIT (EECS, CSAIL, Jameel Clinic)

Associate Professor at MIT in Electrical Engineering and Computer Science, Marzyeh Ghassemi leads the Healthy ML lab, a group that focuses on making machine learning for healthcare robust, private, and fair. Her research has documented how standard ML models trained on healthcare data frequently encode demographic attributes, including race, gender, and insurance status, even when not explicitly trained to do so, leading to systematically unequal performance across patient subgroups. She was named one of MIT Technology Review's 35 Innovators Under 35 and received the 2025 MIT AI Better World Award for her work addressing bias in healthcare AI.

Ghassemi's 2025 research published in Nature Medicine examined how AI models can be tested to ensure they do not inadvertently reveal anonymised patient health data, addressing one of the most underexamined risks in clinical AI deployment. She founded the Association for Health Learning and Inference and serves as a working group lead for the Coalition for Health AI, connecting her academic work directly to the governance conversation. Her commitment to publishing across both computer science and clinical venues ensures that her findings reach both the engineers building AI systems and the clinicians deploying them.

8. Ziad Obermeyer

UC Berkeley School of Public Health

Ziad Obermeyer is the Blue Cross of California Distinguished Professor at UC Berkeley and one of the most influential researchers in healthcare AI, not because he has built the most sophisticated algorithms, but because he has most clearly documented where they fail. His 2019 Science paper, co-authored with Sendhil Mullainathan and colleagues, exposed racial bias in a widely used commercial algorithm affecting 100 million Americans, showing that the algorithm was systematically identifying healthier white patients for care management ahead of sicker Black patients. The work triggered investigations, algorithmic audits, and legislative attention that are still playing out.

Obermeyer followed that landmark paper with the Algorithmic Bias Playbook, developed with colleagues at the University of Chicago Booth School of Business, which provides a practical guide for healthcare organisations to measure, detect, and mitigate bias in the algorithms they deploy. He testified before the US Senate Finance Committee in 2024 and was named to TIME Magazine's 100 Most Influential People in AI. His co-founded company Dandelion Health works to address the data infrastructure problems that allow bias to persist.

9. Pranav Rajpurkar

Harvard Medical School

Associate Professor of Biomedical Informatics at Harvard Medical School, Pranav Rajpurkar is one of the most prolific researchers in AI-powered medical imaging and clinical AI more broadly. He leads a lab developing what he describes as "AI doctors," generalist systems that combine the reasoning capabilities of physicians with the computational power of large-scale models, and his 2025 paper in Nature on multimodal generative AI for medical image interpretation has attracted significant attention. He has published more than 180 papers with over 48,000 citations, including work in Nature Medicine, the New England Journal of Medicine, and Science.

Rajpurkar co-founded a2z Radiology AI, developing the first commercially available multi-finding abdomen-pelvis CT triage AI in the US, bridging the gap between foundational research and clinical deployment in ways that few researchers manage. He educates tens of thousands of practitioners through the AI Health Podcast and the Doctor Penguin newsletter, and his Coursera AI for Medicine series has reached more than 84,000 learners globally.

10. Regina Barzilay

MIT

A Professor at MIT and a cancer survivor herself, Regina Barzilay brings personal conviction to her research on AI-powered cancer screening and treatment. She leads AI-focused oncology research at the MIT Abdul Latif Jameel Clinic for Machine Learning in Health, where she has developed deep learning systems for mammography screening that detect breast cancer years before conventional screening methods. Her work on risk stratification models, which can predict a patient's five-year risk of developing cancer from a single mammogram, has been validated in multi-institutional studies and is being translated into clinical practice in the United States, Sweden, and other countries.

Barzilay received the Squirrel AI Award in 2022, and her work on applying natural language processing to clinical notes to improve oncology care pathways has been adopted by major cancer centres. Her 2025 and 2026 work focuses on how foundation models can be applied to cancer genomics, expanding the reach of her mammography work into the molecular understanding of tumour biology.

11. James Zou

Stanford University

An Associate Professor at Stanford University in the Department of Biomedical Data Science, James Zou builds AI systems for biomedical research and clinical applications with a particular focus on understanding and preventing the failure modes that make AI unsafe in healthcare. His research group has published seminal work on data contamination in large language models, algorithmic fairness in genomics, and the generalisation challenges that prevent AI models trained in one clinical context from performing reliably in another.

Zou's 2025 paper in Nature Biomedical Engineering on coordinated AI agents in healthcare, co-authored with Eric Topol and Pranav Rajpurkar, examines how multiple AI systems working together can take on increasingly complex clinical tasks, and how governance frameworks need to evolve to keep pace with this shift. He co-directs Stanford's Center for Artificial Intelligence in Medicine and Imaging and is recognised as one of the most important emerging voices on both the technical foundations and the clinical translation challenges of health AI.

12. Irene Chen

MIT / UC Berkeley

An Assistant Professor at UC Berkeley in the Department of Electrical Engineering and Computer Sciences, Irene Chen focuses on the fairness, robustness, and reliability of machine learning systems in high-stakes domains, with healthcare as her primary application area. Her research has examined how distribution shifts between training and deployment environments cause AI models to fail in predictable but preventable ways, and how subgroup analysis and model auditing can catch these failures before they reach patients.

Chen's 2025 work on clinical prediction models and the conditions under which they generalise across patient populations has been widely cited by both researchers and clinical informaticists trying to understand which AI tools are safe to deploy beyond their original development contexts. Her commitment to making technical findings accessible to clinical audiences, through both publications and public engagement, makes her one of the most bridging voices in this space.

13. Suchi Saria

Bayesian Health / Johns Hopkins University

Suchi Saria is a Professor at Johns Hopkins University and CEO of Bayesian Health, combining academic rigour with clinical deployment experience in ways that few researchers in this space have managed. Her academic work focuses on applying machine learning to clinical time-series data, developing early warning systems for sepsis, deterioration, and other critical events that allow clinical teams to intervene before a patient's condition becomes life-threatening. Her Targeted Real-Time Early Warning System for sepsis was deployed across multiple health systems and has been associated with measurable reductions in sepsis mortality.

Saria is a board member of the Coalition for Health AI, connecting her deployment experience directly to the governance conversation. Her 2025 and 2026 work examines how agentic AI, systems that can autonomously execute multi-step clinical tasks, should be designed and governed to maintain human oversight at the moments when it matters most. She has been consistently vocal about the difference between AI that is technically impressive and AI that is clinically safe.

14. Leo Celi MD

MIT Lab for Computational Physiology / Harvard Medical School

A Research Scientist at MIT and Lecturer at Harvard Medical School, Leo Celi has spent his career arguing that the most important question in healthcare AI is not whether AI can outperform clinicians in controlled research settings, but whether it can be built and deployed equitably across the full range of health systems globally, including those in low-resource settings that carry the highest burden of preventable disease. He co-leads the MIT Laboratory for Computational Physiology and is one of the founders of MIMIC, the largest publicly available critical care database, which has enabled hundreds of research groups around the world to build and evaluate clinical AI without the access barriers that make health AI research a privilege of wealthy institutions.

Celi's work on datathons, intensive collaborative research events that bring together clinicians, data scientists, and public health researchers to address specific clinical problems, has created a global community of practice that is genuinely international in its composition. His 2025 and 2026 papers continue to focus on representation, fairness, and the conditions under which AI findings from high-income settings can be responsibly applied in low-income ones.

15. Michael Pencina

Duke AI Health

A Professor at Duke University and Director of Duke AI Health, Michael Pencina is one of the leading biostatisticians working on the methodological foundations of clinical AI evaluation. His work addresses a problem that is both technical and consequential: how do you rigorously evaluate whether an AI system actually improves patient outcomes, rather than simply performing well on the metrics it was trained to optimise? He is a board member of the Coalition for Health AI, and his work on calibration, validation, and performance monitoring of clinical prediction models is foundational for anyone building or deploying AI in healthcare.

Pencina has led the development of Duke AI Health into one of the most rigorous institutional programmes for health AI in the United States, combining methodological research with direct engagement with health system leaders on the governance and deployment questions that follow once a model is technically validated. His 2025 work on generative AI in clinical practice examines both the promise and the risks of large language models deployed in clinical decision support.

Category 3: The Clinical AI Translators

These nine people are working at the interface between what AI can do in a research lab and what it can do in a clinical ward. They are the practitioners, clinician-researchers, and health system leaders translating AI from proof-of-concept to clinical reality.

16. Rachael Callcut MD

UC Davis Health

A trauma and acute care surgeon and Vice Chair for Research at UC Davis Health, Rachael Callcut is one of the most important voices on AI in surgical and critical care settings. Her research focuses on applying machine learning to clinical data in the intensive care unit and trauma bay, developing predictive models that give surgical teams earlier warning of complications and deterioration. She has been particularly attentive to the implementation science questions: how do you embed AI into a clinical workflow in ways that clinicians trust, use consistently, and integrate with rather than against their clinical judgment?

Callcut's 2025 and 2026 work addresses the governance and human factors challenges of deploying AI in emergency and surgical settings, where the stakes are highest and the time for deliberation is shortest. She is a visible practitioner-researcher voice at HLTH, ViVE, and other major health technology conferences, and her LinkedIn posts are grounded in the practical realities of clinical AI deployment in complex, resource-constrained settings.

17. Mark Sendak MD

Duke Institute for Health Innovation

A physician and Co-Founder of the Duke Institute for Health Innovation, Mark Sendak leads one of the most respected programmes in the United States for translating AI research into clinical deployment. His work focuses on the specific institutional challenges that arise when AI moves from a research prototype to a system running in production on real patients: the governance questions, the monitoring frameworks, the failure modes that only become visible at scale, and the human change management work required to make clinical teams actually use a tool that is technically sound.

Sendak published foundational work on the lifecycle of AI in clinical practice, documenting the gap between research performance and real-world deployment performance that consistently undermines AI's promise in healthcare. His 2025 papers on clinical AI governance and the conditions for safe deployment have become standard references in health system AI programmes. He posts regularly on LinkedIn about the implementation realities that academic literature underreports.

18. David Bates MD

Brigham and Women's Hospital / Harvard Medical School

A Professor of Medicine at Harvard Medical School and Chief of General Internal Medicine at Brigham and Women's Hospital, David Bates is one of the founders of patient safety science and has been applying those principles to healthcare AI for more than a decade. He is perhaps best known for his pioneering work on computerised physician order entry and clinical decision support, research that established both the promise and the unintended consequences of decision support systems in ways that directly inform how AI should be designed and deployed today.

Bates has been one of the most consistent voices arguing that AI in healthcare should be evaluated not just for accuracy in controlled settings but for its impact on clinical outcomes and safety in practice. His 2025 work on AI in ambulatory care settings and the governance challenges of deploying AI in primary care practices without the resources of large health systems is particularly important given how much health AI research focuses on tertiary academic centres.

19. Isaac Kohane MD

Harvard Medical School

Chair of the Department of Biomedical Informatics at Harvard Medical School and Editor-in-Chief of NEJM AI, Isaac Kohane occupies a unique position in the healthcare AI ecosystem: he is simultaneously a senior academic researcher, a practising physician, a journal editor shaping what counts as rigorous evidence in this field, and a vocal critic of the conditions under which AI is being deployed at a pace that outstrips the evidence for its safety and efficacy. His 2023 book The AI Revolution in Medicine, co-authored with Peter Lee and Carey Goldberg, was one of the first accessible accounts of what large language models might actually do in clinical practice, written with the intellectual honesty to acknowledge both their promise and their current limitations.

Kohane has been a consistent voice for rigour in a field that sometimes moves faster than its evidence base. His work at NEJM AI, selecting and editing the studies that will define what rigorous health AI research looks like, makes him one of the most influential arbiters of quality in this space. His LinkedIn posts combine deep technical knowledge with a clinician's perspective on what these systems need to do before they are ready for practice.

20. Sara Murray MD

UCSF

A clinical informaticist and practising physician at the University of California San Francisco, Sara Murray focuses on how AI and information technology can improve clinical workflows and reduce administrative burden on clinicians. Her work spans both clinical decision support and the governance infrastructure that health systems need to evaluate and safely deploy AI tools. She is a member of the AMIA board and has been active in building the professional standards for clinical informatics that bridge medicine and technology.

Murray's 2025 and 2026 work addresses the specific challenge of AI in outpatient and primary care settings, where implementation resources are more limited than in large academic medical centres but the potential to improve access and quality is significant. Her LinkedIn presence is consistently substantive, addressing both the promise of AI in reducing clinician burnout and the conditions that need to be in place for that promise to be realised safely.

21. Harlan Krumholz MD

Yale School of Medicine

A Professor of Medicine and Director of the Yale New Haven Hospital Center for Outcomes Research and Evaluation (CORE), Harlan Krumholz is one of the most important voices on evidence generation and data transparency in medicine, principles he has been applying systematically to healthcare AI. His research group studies the conditions under which AI can genuinely improve cardiovascular outcomes, and he has been consistently critical of the gap between impressive model performance metrics and actual evidence of clinical benefit.

Krumholz co-founded the Open Science Alliance and has been a vocal advocate for data sharing in medicine, arguing that the proprietary data silos that currently structure health AI development create conditions for bias and error that cannot be detected or corrected without transparency. His work on data transparency directly challenges the business model of most commercial health AI vendors, which is what makes his voice important and sometimes uncomfortable in industry-sponsored settings.

22. Raina Merchant MD

Penn Medicine / University of Pennsylvania

A Professor in the Department of Emergency Medicine at the University of Pennsylvania and Director of the Penn Medicine Center for Digital Health, Raina Merchant is one of the most important researchers on AI in social media and digital health. Her work has demonstrated that patterns in social media data can predict clinical events including cardiac arrest and other time-sensitive conditions, opening new possibilities for population-level health surveillance that has direct implications for how AI can be deployed in public health and emergency response.

Merchant was quoted alongside colleagues in the New York Times in 2025 in a piece on how AI is changing clinical practice, reflecting her growing public profile as a voice who can translate complex health AI research for general audiences. Her 2025 and 2026 work continues to explore how digital data from consumer platforms can be used for clinical and public health applications, with careful attention to the consent, equity, and accuracy questions that this raises.

23. Danielle Bitterman

Dana-Farber Cancer Institute / Harvard Medical School

An instructor at Harvard Medical School and researcher at Dana-Farber Cancer Institute, Danielle Bitterman focuses on natural language processing and large language models in oncology, with particular attention to how these tools can be used to improve communication between patients and clinical teams, reduce administrative burden on oncologists, and support the complex documentation requirements of cancer care. Her research on AI-generated clinical documentation and patient communication is at the frontier of a rapidly evolving applied AI space.

Bitterman's 2025 work on the KScope framework, developed with Marzyeh Ghassemi's lab at MIT and published in November 2025, examines how to characterise the knowledge status of large language models, a methodological contribution with direct implications for when and how LLMs can be trusted in clinical settings. Her focus on the intersection of cancer care, patient experience, and AI makes her one of the distinctive voices connecting technical AI research to the human realities of serious illness.

Category 4: The Equity and Ethics Voices

These seven people focus on the justice dimensions of healthcare AI, asking not just whether AI works but whether it works for everyone, and what the conditions are under which it might make existing health disparities worse rather than better.

24. Judy Gichoya MD

Emory University

An Associate Professor in Radiology and Imaging Sciences at Emory University and co-director of the Healthcare AI Innovation and Translational Informatics (HITI) lab, Judy Gichoya is one of the most important voices globally on bias in medical AI. Her 2022 paper in The Lancet Digital Health, showing that AI can identify a patient's race from medical images with high accuracy even when not trained to do so, triggered a field-wide reassessment of what information AI systems are encoding and how that encoding might be used in harmful ways. The paper set off what she describes as a national firestorm, one that is still reshaping how the field thinks about fairness in clinical AI.

Gichoya leads the annual HITI Datathon, which brings together researchers from around the world to build diverse datasets and develop bias-detection techniques for clinical AI. She has been recognised as a National Academy of Medicine Emerging Leader and ranked among the top 50 most influential figures in radiology. Her 2026 paper on rethinking human-AI collaboration in radiology, published in Radiology, examines the human factors and institutional conditions that determine whether AI helps or hinders radiologists in practice.

25. Adewole Adamson MD

University of Texas at Austin

An Assistant Professor of Dermatology at the University of Texas at Austin, Adewole Adamson is one of the leading voices on racial and demographic bias in dermatology AI, a field where the consequences of training AI on unrepresentative datasets are particularly visible and well-documented. His research has shown that most commercially available AI tools for skin condition diagnosis perform significantly worse on darker skin tones, because the datasets used to train them are heavily skewed toward lighter-skinned patients.

Adamson has published in leading dermatology and AI journals and testified before the FDA about the regulatory frameworks needed to ensure AI-powered diagnostics are validated for the full diversity of patients who will use them. His 2025 work on dataset diversity standards and the clinical validation requirements for AI in dermatology is increasingly cited by regulators, vendors, and health systems seeking to understand what responsible AI deployment in this specialty actually requires.

26. Kadija Ferryman

Johns Hopkins University

A Core Faculty member at the Johns Hopkins Berman Institute of Bioethics and Assistant Professor in the Department of Health Policy and Management at the Johns Hopkins Bloomberg School of Public Health, Kadija Ferryman approaches healthcare AI from the disciplines of anthropology and critical science and technology studies, bringing a methodological perspective that is rare in a field dominated by engineers and clinicians. Her work examines how social and structural factors shape the design, deployment, and impact of health technologies, and why well-intentioned AI systems so often reproduce the inequities of the health systems they are designed to improve.

Ferryman's work on algorithmic fairness frameworks in healthcare examines the gap between how fairness is defined in machine learning research and how equity is understood in public health and social medicine, arguing that AI governance frameworks built on technical fairness metrics alone will systematically fail to address structural health disparities. Her ability to make social science perspectives legible to technical audiences, and vice versa, makes her a bridging voice the field genuinely needs. Her research has been featured in Nature, STAT, the New England Journal of Medicine, and The Financial Times.

27. Ravi Parikh MD

University of Pennsylvania

An Assistant Professor of Medical Ethics and Health Policy at the University of Pennsylvania, Ravi Parikh focuses on how AI interacts with clinical decision-making in oncology and how the design of AI systems shapes the choices that clinicians and patients make. His research has examined how clinical decision support tools, including those powered by AI, can create automation bias, the tendency of clinicians to over-rely on algorithmic recommendations even when their own clinical judgment should override them. He has published extensively on how AI governance needs to account for these human factors effects to avoid creating new categories of preventable harm.

Parikh's 2025 work on AI and end-of-life care decision-making addresses one of the most sensitive frontiers in clinical AI: the application of predictive models to decisions about prognosis, goals of care, and hospice referral. His ethical analysis of these use cases, and his advocacy for governance frameworks that protect patient autonomy, represents an important counterweight to purely technical framings of the same questions.

28. Eleni Linos MD

Harvard Chan School of Public Health

A Professor and founding Director of the Program in AI, Law, and Society at Harvard Chan School of Public Health, Eleni Linos focuses on the intersection of dermatology, AI, and health equity, with particular attention to how AI might be used to expand access to dermatological care in settings where specialists are scarce. Her research combines clinical and public health perspectives, examining both the technical performance of skin disease AI and the equity implications of how and where it is deployed.

Linos's 2025 work on AI in skin cancer screening programmes in under-resourced settings examines the conditions under which AI-assisted screening can genuinely reduce disparities rather than simply shifting diagnostic capacity without addressing the underlying access barriers. She is one of the few researchers in this space combining rigorous technical evaluation with public health equity analysis.

29. Stacy Lindau MD

University of Chicago

A Professor in the Departments of Obstetrics and Gynecology and Medicine at the University of Chicago, Stacy Lindau is the founder of NowPow and one of the most innovative voices on social determinants of health and how AI can connect patients to the community resources that address the non-clinical drivers of health outcomes. Her work on community health and information technology has shaped how health systems think about the interface between clinical care and social care, a dimension of healthcare AI that most technical research ignores entirely.

Lindau's foundational contribution is demonstrating that access to community resources, including housing, food, transportation, and social connection, is as important to health outcomes as clinical care, and that AI can play a role in systematically connecting patients to those resources in ways that primary care practices cannot currently do manually. Her 2025 and 2026 work continues to examine how digital tools can bridge the clinical and social care systems in ways that reduce disparities and improve outcomes for the most vulnerable patients.

30. Lucila Ohno-Machado

Yale School of Medicine

A Professor and founding Chair of the Department of Biomedical Informatics and Data Science at Yale School of Medicine, Lucila Ohno-Machado is one of the most important figures in health data governance, informatics standards, and the responsible use of clinical data for AI research. She has led major national initiatives on privacy-preserving data sharing, federated learning in healthcare, and the development of common data models that allow AI research to happen across institutions without centralising sensitive patient data.

Her work on privacy-preserving methods for clinical AI is foundational for anyone thinking seriously about how to build AI that can learn from population-scale data without creating the surveillance and discrimination risks that uncontrolled data sharing would generate. Her move to Yale brought her to one of the most active health AI research environments in the US, and her 2025 and 2026 work continues to advance the infrastructure for responsible, privacy-preserving health AI at scale.

Category 5: The Global and International Voices

These eight people bring perspectives from outside the US-UK corridor that dominates most health AI discourse. They are working on health AI challenges in Africa, Australia, the Middle East, Europe, and Asia, and their work is essential for understanding what responsible global health AI actually requires.

31. Enrico Coiera

Macquarie University, Australia

Professor Enrico Coiera is the Foundation Professor in Medical Informatics at Macquarie University and founding Director of the Centre for Health Informatics, Australia's largest academic research group in digital health. Trained in both medicine and computer science with a PhD in AI, he is one of the longest-serving and most internationally recognised researchers in health informatics. His textbook Guide to Health Informatics is in its third edition and is used in health informatics programmes across more than 30 countries.

Coiera's 2026 paper in JMIR Medical Informatics on AI scribes, asking whether the field is measuring what matters in the evaluation of ambient documentation tools, is a timely critique of one of the most rapidly adopted categories of clinical AI. His contribution to Australia's National Policy Roadmap for AI in Healthcare has shaped the country's approach to governing AI deployment in clinical settings. He consistently argues that clinical adoption of AI scribes has outpaced the evidence for their safety and efficacy, and that the field needs better evaluation frameworks before celebrating deployment as success.

32. Dympna Hannigan

Trinity College Dublin

A Professor in Health Informatics at Trinity College Dublin and founding Director of the Health Informatics Research Group, Dympna Hannigan is one of Europe's most important voices on digital health implementation and the conditions under which health information technology, including AI, actually improves patient outcomes rather than simply adding complexity to clinical workflows. Her research focuses on the human and organisational factors that determine whether digital health investments succeed or fail, from electronic health record implementation to AI-powered clinical decision support.

Hannigan's work on digital health equity in Ireland and Europe examines how the gap between institutions with robust digital infrastructure and those without is shaping the distribution of benefits from health AI, in ways that risk entrenching rather than reducing existing health inequalities. Her 2025 research on implementation science in digital health provides frameworks that health system leaders need to use before and during AI deployment, not as an afterthought.

33. Bertalan Mesko MD

The Medical Futurist Institute

The Director of The Medical Futurist Institute and a Private Professor at Semmelweis Medical School in Budapest, Bertalan Mesko has been one of the most prolific global educators on healthcare technology and AI since 2011. As an Amazon Top 100 author with a PhD in genomics, he has delivered more than 900 keynotes at Harvard, Stanford, Yale, the WHO, and the world's leading pharmaceutical companies, building a platform that makes the future of healthcare technology accessible to clinical, executive, and policy audiences.

Mesko's 2026 book Your Map to the Future is his most comprehensive synthesis of how AI, genomics, wearable technology, and digital health tools will reshape care delivery over the coming decade, written in the tradition of rigorous futures analysis rather than uncritical technology enthusiasm. He posts extensively on LinkedIn about the week's most significant developments in digital health and AI, and his consistent framing of technology in terms of what it means for the patient-clinician relationship makes him one of the most humanising voices in a field that can become very technical very quickly.

34. Ara Darzi

Imperial College London

A Professor at Imperial College London, a practising colorectal surgeon, and a member of the House of Lords, Ara Darzi is one of the most influential voices at the intersection of surgical innovation, health system design, and AI governance in the United Kingdom. He led the landmark Darzi Review, which shaped NHS reforms under multiple UK governments, and has been a consistent advocate for the application of AI and digital innovation in healthcare in ways that keep the patient experience at the centre.

His 2025 work through the Institute of Global Health Innovation at Imperial continues to examine how AI can improve surgical outcomes and how health systems need to be designed to support responsible AI adoption. As both a practising clinician and a policy influencer, Darzi occupies a rare position of being able to shape both the technical direction and the governance framework of health AI in a major national health system.

35. Hutan Ashrafian

Imperial College London

A Reader at Imperial College London and a practising cardiac surgeon, Hutan Ashrafian is one of the most prolific researchers on AI in surgery and the history and ethics of AI in medicine more broadly. He has published extensively on surgical robotics, AI-assisted diagnosis, and the governance frameworks needed to regulate AI in high-risk clinical settings. His research on the ethical dimensions of AI in medicine is grounded in clinical practice rather than abstract philosophy, making it directly relevant to the practitioners and policymakers who need to make decisions about AI deployment.

Ashrafian's 2025 work on AI and surgical decision-making examines the automation bias risks in surgical AI, the conditions under which algorithmic recommendations might override surgical judgment in ways that harm patients, and the design principles that could mitigate those risks. He is an active LinkedIn contributor whose posts combine clinical perspective with policy insight.

36. Jan Hazelzet MD

Erasmus MC

A Professor and Chief Medical Information Officer at Erasmus MC in Rotterdam, Netherlands, Jan Hazelzet is one of Europe's leading voices on clinical informatics, AI in intensive care, and the governance of health data. His work focuses on how electronic health record data can be used to train AI systems that improve clinical decision-making in intensive care and paediatrics, and on the data governance frameworks needed to make that happen responsibly at scale.

Hazelzet has been instrumental in building the international MIMIC and eICU research communities in Europe, and his work on federated learning approaches that allow AI training across multiple European health systems without centralising data addresses one of the most significant barriers to European health AI development. His 2025 and 2026 papers on clinical AI governance frameworks for European health systems are particularly relevant given the EU AI Act's implications for regulated medical AI products.

37. Tova Patalon MD

Maccabi Healthcare Services

The Director of the Maccabi Research and Innovation Center at Maccabi Healthcare Services in Israel, Tova Patalon is one of the world's leading practitioners of population-scale health data analysis and AI-powered public health surveillance. Maccabi serves approximately 2.7 million members across Israel and maintains one of the most comprehensive longitudinal health databases in the world, which Patalon's team has used to produce foundational research on COVID-19 vaccine effectiveness, chronic disease management, and AI-powered early warning systems.

Her 2025 and 2026 work on agentic AI in healthcare examines how autonomous AI systems can be used to support population health management at scale while maintaining the governance structures needed to ensure patient safety. Her experience deploying AI in a real-world, population-level health system distinguishes her perspective from most academic researchers in this space.

38. Kira Radinsky

Diagnostic Robotics / Technion

Kira Radinsky is the CTO and scientific co-founder of Diagnostic Robotics, an AI-powered care management platform, and a Visiting Professor at the Technion in Israel. She is one of the most accomplished entrepreneurial researchers in health AI, with a career that has spanned data science at eBay, the co-founding of SalesPredict (acquired by eBay in 2016), and the building of Diagnostic Robotics into a platform used by major US health systems and payers. She was named to MIT Technology Review's 35 Innovators Under 35 and recognised by Forbes as one of the 30 Under 30 Rising Stars in Enterprise Technology.

Radinsky's clinical AI work focuses on predicting patient deterioration, care gaps, and utilisation patterns at the population level, and her platform's integration with major US payers and health systems gives her deployment experience at a scale that few academic researchers match. Her recent work on AI-driven care management and prior authorisation is at one of the most consequential frontiers in US healthcare AI.

39. Alastair van Heerden

Wits Health Consortium

A Research Director at the Wits Health Consortium at the University of the Witwatersrand in South Africa, Alastair van Heerden is one of the most important voices on AI and mHealth in sub-Saharan Africa. His work focuses on how mobile health technologies and AI can improve mental health, HIV, and maternal and child health outcomes in resource-constrained settings, addressing the specific challenges of deploying digital health tools where infrastructure is limited, literacy varies, and the health burdens are highest globally.

Van Heerden's February 2026 paper in the Transactions of the Royal Society of Tropical Medicine and Hygiene on a multi-agent AI chatbot for HIV and mental health support in South Africa is a concrete example of how AI can be designed for the contexts that most health AI research ignores. He received the UCSF Edison T. Uno Award for Public Service in 2025 for his contributions to global health equity through technology.

Category 6: The Radiology and Imaging AI Specialists

These six people are working at the frontier where AI has achieved its most clinically demonstrable results: medical imaging. Their work spans the development of algorithms, the validation of their clinical performance, and the governance questions that arise as these tools move into routine use.

40. Tina Hernandez-Boussard

Stanford University

A Professor at Stanford University in the Departments of Medicine, Biomedical Data Science, and Epidemiology, Tina Hernandez-Boussard is one of the leading researchers on AI implementation and evaluation in clinical practice, with particular expertise in surgical AI and the use of electronic health record data to develop and validate clinical prediction models. Her research focuses on how AI can reduce clinical practice variation in surgery and improve the equity of surgical outcomes across patient populations.

Hernandez-Boussard's 2025 work on the conditions under which AI-generated clinical documentation improves or degrades the quality of surgical care pathways examines a question that is both technically and clinically important: when ambient AI documentation creates efficiencies in the short term, what are the downstream effects on clinical reasoning and documentation quality? Her Stanford affiliation gives her access to one of the richest clinical data environments in the world for this kind of research.

41. Roxana Daneshjou MD

Stanford University

An Assistant Professor of Biomedical Data Science and Dermatology at Stanford University, Roxana Daneshjou is one of the leading voices on AI in dermatology and on the diversity and representation challenges in medical imaging AI more broadly. Her work has demonstrated how commercial AI tools for skin disease diagnosis perform significantly worse on skin conditions as they appear on darker skin tones, and she has been a consistent advocate for dataset diversity standards and inclusive validation requirements for clinical AI in dermatology and beyond.

Daneshjou's 2025 research on large language models in clinical dermatology examines both the capabilities and the limitations of foundation models for clinical use, with particular attention to how they handle rare conditions and presentations that are underrepresented in their training data. She posts actively on LinkedIn about the practical realities of AI in clinical practice, bridging research findings and clinical experience in ways that both communities find valuable.

42. Constance Lehman MD

Massachusetts General Hospital / Harvard Medical School

A Professor of Radiology at Harvard Medical School and Chief of Breast Imaging at Massachusetts General Hospital, Constance Lehman is one of the world's leading authorities on AI in breast imaging and mammography screening. Her research on deep learning algorithms for mammography interpretation has produced some of the strongest evidence in the literature for the clinical utility of AI in radiology, demonstrating reductions in both false-positive and false-negative rates when AI is integrated appropriately into screening workflows.

Lehman's work is notable for its methodological rigour: she has been consistently careful to study AI performance in the specific populations and clinical contexts where it will be deployed, rather than relying on aggregate accuracy metrics that may obscure performance differences across subgroups. Her 2025 and 2026 papers continue to advance the evidence base for AI in breast imaging and to address the governance and implementation questions that arise as AI screening tools move into routine clinical practice.

43. Aashima Gupta

Google Cloud Healthcare

The Global Director of Healthcare Solutions at Google Cloud, Aashima Gupta is one of the most influential technology executives shaping how cloud infrastructure and AI are being adopted by health systems globally. She is a member of the HIMSS Board of Advisers and leads Google Cloud's work on applying cloud computing, AI, APIs, and mobile solutions to accelerate healthcare's digital transformation. Her role gives her visibility into AI adoption patterns across hundreds of health systems simultaneously, providing a perspective on what is actually working in deployment that complements the academic research literature.

Gupta is a consistent advocate for health equity in AI, and her work at Google Cloud focuses on building the infrastructure standards and interoperability frameworks that allow AI to work across diverse health system contexts. She posts regularly on LinkedIn about AI and digital health transformation, and her ability to articulate the healthcare technology perspective to both technical and clinical audiences makes her a bridge voice between these communities.

44. Nigam Shah

Stanford University / Stanford Health Care

A Professor of Medicine and Chief Data Scientist at Stanford Health Care, Nigam Shah is one of the world's leading researchers in clinical natural language processing and the use of electronic health record data for machine learning. His work has produced both foundational research in clinical NLP and practical tools that have been adopted by health systems to extract clinical insights from unstructured documentation, a category of information that has historically been invisible to data analysis. He co-founded Atropos Health, which provides clinicians with on-demand patient outcome summaries based on real-world data.

Shah's 2025 work on the transition from clinical AI as a tool for individual decision support to clinical AI as infrastructure for population health management examines how the governance frameworks and validation requirements need to evolve as AI becomes more deeply embedded in clinical workflows. His dual role at Stanford as both an academic researcher and Chief Data Scientist for the health system gives him a perspective that combines rigour with real-world deployment accountability.

45. Hamid Tizhoosh

Mayo Clinic / University of Waterloo

A researcher at Mayo Clinic and Professor at the University of Waterloo, Hamid Tizhoosh is one of the leading researchers in computational pathology and AI-powered tissue analysis. His work focuses on developing AI systems that can analyse whole-slide pathology images at a level of speed and consistency that humans cannot match, identifying patterns and features that correlate with prognosis, treatment response, and disease progression in cancer and other conditions.

Tizhoosh has developed search-based approaches to pathology AI, systems that can retrieve the most similar cases from a large archive when analysing a new slide, providing pathologists with contextual information that improves diagnostic accuracy and consistency. His 2025 and 2026 work on foundation models for pathology examines how large AI systems trained on millions of pathology images can be adapted for specific diagnostic tasks with relatively small amounts of task-specific training data.

Category 7: The Digital Health Educators and Future-Builders

These seven people are shaping how the next generation of clinicians, researchers, and leaders understand and navigate AI in healthcare. They build communities, educate practitioners, produce the books and podcasts and teaching programmes that make this complex field accessible.

46. Wynne Hsu

National University of Singapore

A Professor in the Department of Computer Science at the National University of Singapore and co-Director of N.1 Institute for Health, Wynne Hsu is one of the leading researchers in Asia on AI-powered clinical decision support and health informatics. Her work spans natural language processing for clinical documentation, predictive modelling for chronic disease management, and the governance frameworks needed to deploy AI safely in diverse healthcare settings across Southeast Asia.

Hsu's contribution to the Asian health AI ecosystem is significant: she has built research programmes that address the specific data, infrastructure, and clinical workflow challenges of healthcare systems in Singapore and neighbouring countries, which are different in important ways from the US and European contexts that dominate most health AI research. Her 2025 work on AI for early detection of chronic kidney disease in primary care settings in Singapore provides a model for how AI can be deployed equitably in contexts where preventive care capacity is strong but specialist referral pathways are limited.

47. Charlene Liew

Nanyang Technological University

An Assistant Professor at Nanyang Technological University's Lee Kong Chian School of Medicine in Singapore, Charlene Liew focuses on clinical AI applications in radiology and oncology, with particular attention to how AI can improve cancer screening and early detection in Asian populations whose disease risk profiles differ from the Western populations that most AI training datasets predominantly represent. Her work addresses both the technical challenges of developing AI for diverse populations and the implementation challenges of deploying it in healthcare systems with different infrastructure, regulation, and clinical workflow assumptions.

Liew's 2025 and 2026 research on multimodal AI for lung cancer screening and liver cancer surveillance in Asian populations is filling a genuine evidence gap, and her collaboration with clinical teams across Singapore and the broader Asia-Pacific region gives her work translational relevance that connects laboratory research to clinical practice.

48. Alejandro Jadad MD

University of Toronto

A Professor at the University of Toronto and founder of multiple digital health initiatives, Alejandro Jadad is one of the world's leading researchers on digital health and patient empowerment. His landmark work on evidence-based medicine in the digital age, including the development of the Jadad Scale for assessing the quality of randomised controlled trials, was foundational for how the field thinks about information quality and evidence standards in a world where patients have access to vastly more health information than was previously imaginable.

Jadad's current work focuses on how AI can be used to support patients in managing their own health and navigating complex healthcare systems, and on the governance frameworks needed to ensure that patient-facing AI tools are accurate, equitable, and genuinely empowering rather than creating new forms of digital health inequality. His 2025 work on AI and patient autonomy examines the conditions under which AI decision support for patients augments rather than undermines their capacity for informed self-determination.

49. Melek Somai

George Washington University

An Assistant Research Professor at George Washington University's School of Medicine and Health Sciences and a physician, Melek Somai focuses on the intersection of digital health, AI, and health policy, with particular attention to how AI tools need to be designed and regulated to support rather than displace the patient-clinician relationship. Her research addresses the consent, transparency, and accountability questions that arise when AI is used to inform clinical decisions, and she has been a visible voice at health policy and technology conferences on the governance questions that regulators and health systems are still working through.

Somai's 2025 work on AI and consumer wearable health data examines how the explosion of consumer health monitoring technologies, from smartwatches to continuous glucose monitors, is creating both new clinical possibilities and new governance challenges around data ownership, consent, and the clinical responsibility for acting on information that patients increasingly arrive with before their clinical appointments.

50. Jonno White

Consult Clarity

The people on this list are the thinkers. Jonno White is the person organisations bring in when they are ready to act on what they say. A Certified Working Genius Facilitator and bestselling author of Step Up or Step Out with over 10,000 copies sold globally, Jonno helps healthcare leadership teams build the trust, clarity, communication, and alignment that translate AI strategy into organisational practice. The hardest part of deploying AI in healthcare is rarely the algorithm. It is the human system: the leadership team that needs to make aligned decisions, the clinical culture that needs to shift, the difficult conversations that need to happen between clinicians, executives, and boards before change becomes real. Jonno's keynotes, workshops, and executive facilitation work help healthcare organisations close that gap.

Working with hospitals, health networks, and healthcare associations across Australia and internationally, Jonno delivers the facilitation and leadership development work that turns AI strategy into Monday morning action. As host of The Leadership Conversations Podcast, with more than 230 episodes reaching listeners in 150+ countries, Jonno brings a global perspective on the leadership challenges that make or break major organisational change. International travel is often far more affordable than organisations expect. To bring Jonno in for a keynote, workshop, or executive offsite, email jonno@consultclarity.org.

Notable Voices We Almost Included

Several thinkers were seriously considered but did not make the final list. Eric Topol of Scripps Research is perhaps the most cited physician-researcher in the field, with his books Deep Medicine and The Patient Will See You Now shaping the conversation about AI in healthcare for a decade. John Halamka of Mayo Clinic Platform has driven some of the most important practical AI deployments in health systems globally. Brene Brown, Adam Grant, and Simon Sinek would appear on most lists about leadership or organisational change. Their work has shaped how many organisations think about culture and communication. We deliberately moved past these household names to surface voices the reader may not yet have encountered.

Others who were considered include Atul Butte of UCSF, whose work on genomics and clinical data science is foundational, and Robert Pearl of Stanford, whose writing on AI and physician practice is consistently clear and provocative. Khaled El Emam of the University of Ottawa was considered for his work on privacy-preserving AI methods. All were strong candidates. The final list prioritised geographic diversity and mid-career researchers whose current output is particularly active.

Common Mistakes to Avoid

The most common mistake healthcare leaders make when engaging with AI thought leadership is treating it as a technology conversation rather than a governance and culture conversation. The researchers on this list have spent careers documenting that the hardest problems in health AI are not algorithmic but institutional: the incentive structures, data governance frameworks, clinical workflow assumptions, and equity considerations that determine whether a technically sound AI system produces good outcomes in practice. If you are following only the engineers building these systems and ignoring the clinicians, ethicists, and equity researchers evaluating them, your understanding of the field is dangerously incomplete.

A second common mistake is treating US-based research as universally applicable. Healthcare systems differ profoundly in their data infrastructure, regulatory environments, clinical workflow assumptions, and patient population diversity. AI tools developed and validated in large US academic medical centres frequently underperform in community hospitals, rural settings, safety-net providers, and healthcare systems outside the US. The global voices on this list, from Australia, Ireland, Israel, Singapore, South Africa, the Netherlands, and elsewhere, are not supplementary to the main conversation. They are essential for understanding what responsible global health AI actually requires.

A third mistake is following only the optimists. The field has no shortage of voices predicting that AI will transform healthcare within a decade. What it needs more of are the voices asking under what conditions that transformation will benefit all patients rather than only the most privileged. The equity and ethics voices on this list represent that perspective, and engaging seriously with their work is not optional for anyone making deployment decisions that will affect patients.

Implementation Guide: Taking Action

Start with the governance voices. Before you can evaluate any specific AI application, you need a framework for asking the right questions. Brian Anderson's CHAI materials, Sherri Douville's work on standards operationalisation, and Jonah Feldman's analysis of the legislative landscape will give you that framework. Read or follow at least two of these three before diving into the technical literature.

Then engage with the equity researchers. Ziad Obermeyer's Algorithmic Bias Playbook, Judy Gichoya's papers on bias in medical imaging, and Marzyeh Ghassemi's research on fairness in ML for health will give you the critical lens that prevents you from accepting vendor claims at face value. These are not niche concerns. They are central to whether AI deployments produce good clinical outcomes.

Follow at least one global voice from outside the US-UK corridor. The healthcare system contexts that Enrico Coiera documents in Australia, Alastair van Heerden in South Africa, and Jan Hazelzet in the Netherlands provide reference points that will help you evaluate which findings from US academic medical centres are generalisable and which are specific to that context.

Subscribe to NEJM AI, the journal Isaac Kohane edits, and to the Doctor Penguin newsletter that Pranav Rajpurkar co-edits. Both publish at the frontier of the field and maintain editorial standards that filter out the noise. Set up updates for CHAI, the Coalition for Health AI, to stay current on governance developments without having to actively track multiple sources.

Finally, attend or follow at least one major conference: HIMSS for health IT implementation, HLTH for the business and startup ecosystem, and the ACM Conference on Health, Inference and Learning (CHIL) for the research frontier. These events are where the people on this list present their newest work and where the most important debates in the field are happening in real time.

Jonno White helps healthcare leadership teams build the alignment, communication, and decision-making culture that makes complex AI strategy translatable into clinical practice. Many organisations find that international travel for facilitation is far more affordable than expected. Email jonno@consultclarity.org.

Frequently Asked Questions

What is the current state of AI in healthcare in 2026?

AI in healthcare has moved decisively from experimentation to deployment, but the transition is uneven. According to Menlo Ventures' 2025 State of AI in Healthcare report, surveying more than 700 US healthcare executives, 22% of healthcare organisations have now implemented domain-specific AI tools, a sevenfold increase over 2024 and tenfold over 2023. Ambient documentation has emerged as the first breakout clinical AI category, with Abridge, Ambience, and Nuance's DAX Copilot collectively generating more than USD 600 million in 2025. However, clinical AI for diagnosis and treatment, as opposed to administrative workflows, remains earlier in its adoption curve, with significant governance, equity, and evidence questions still being resolved.

Why do healthcare AI systems sometimes perform differently across patient populations?

AI systems learn patterns from the data they are trained on. If that training data does not represent the full diversity of patients who will use the system in deployment, the system will perform less well for underrepresented groups. Ziad Obermeyer's foundational research documented how a widely used risk-stratification algorithm dramatically underidentified Black patients for care management, because it used healthcare cost as a proxy for need in a system where structural barriers meant Black patients generated lower costs at equivalent clinical severity. This problem affects imaging AI, clinical decision support, and any system where training data reflects historical patterns of access, treatment, or documentation that are themselves shaped by structural inequity.

What is the Coalition for Health AI and why does it matter?

The Coalition for Health AI, founded by Brian Anderson MD in 2021, is a non-profit coalition developing consensus-driven guidelines and independent quality assurance frameworks for AI in healthcare. It has grown to include more than 1,000 active workgroup members and has established voluntary AI governance standards in partnership with The Joint Commission. In a field where FDA regulation covers only a subset of clinical AI tools and where most health systems lack the internal expertise to evaluate vendor claims independently, CHAI is attempting to build the trust infrastructure that responsible AI adoption requires. Its 2025 Impact Report documents significant progress on governance frameworks, assurance lab development, and equity standards.

How is AI changing radiology specifically?

Radiology has been the leading edge of clinical AI adoption, and the evidence for AI's utility in specific radiology tasks is stronger than in almost any other clinical domain. AI systems can now detect time-sensitive conditions including intracranial haemorrhage, pulmonary embolism, and cervical spine fractures with accuracy comparable to expert radiologists, and AI-powered mammography screening tools have demonstrated both sensitivity and specificity improvements in large prospective trials. By 2026, the FDA has cleared nearly 1,000 AI-enabled medical devices, with radiology AI representing the largest share. The remaining governance and implementation challenges are real: automation bias, performance gaps across patient subgroups, and the integration with clinical workflows and specialist reporting pathways all require ongoing attention.

What are the biggest risks of AI in healthcare?

The most important risks are equity risks: AI systems that perform differently across patient subgroups can scale disparities rather than reduce them. Algorithmic bias in screening, risk-stratification, and clinical decision support can affect millions of patients simultaneously in ways that no individual clinical error could replicate. Governance risks are also significant: the pace of deployment has outpaced both regulatory oversight and institutional governance capacity in most health systems, creating conditions for preventable harm. Automation bias, the tendency of clinicians to over-rely on algorithmic recommendations, is a clinical safety risk that is being documented in radiology and other AI-assisted specialties. And the concentration of AI capability in large technology vendors and wealthy health systems risks deepening the digital divide between well-resourced and under-resourced healthcare settings.

Can I hire someone to facilitate AI strategy and leadership workshops for my healthcare team?

Yes. Jonno White is a Certified Working Genius Facilitator, trusted facilitator with healthcare organisations around the world, and the bestselling author of Step Up or Step Out. He works with hospital leadership teams, health networks, and healthcare associations to build the communication, alignment, and decision-making culture that allows AI strategy to translate into clinical practice. To book Jonno for a keynote, workshop, or executive offsite, email jonno@consultclarity.org. Many healthcare organisations find that international travel for facilitation is far more affordable than expected.

What should healthcare leaders do first when thinking about AI adoption?

The most important first step is governance, not technology selection. Before selecting an AI tool, health system leaders need to establish who is responsible for evaluating AI performance, how bias and equity will be assessed, what monitoring frameworks will be in place after deployment, and how clinicians will be trained to understand both the capabilities and the limitations of the tools they are using. The CHAI Responsible Use of AI in Healthcare guidance, developed jointly with The Joint Commission, provides a practical starting point for health systems at any level of AI maturity.

How was this list compiled?

This list prioritises genuine contribution to the field over institutional prestige. Candidates were assessed on their published work, active contributions to the field in 2025 and 2026, geographic diversity, disciplinary diversity, and their specific impact on how the field thinks about AI governance, equity, clinical deployment, and global health. The process deliberately sought out mid-career and emerging voices who are actively shaping current discourse, while also including established researchers whose work is foundational. No single country represents more than 35% of the final list, with representation from North America, the United Kingdom, Europe, Australia, Israel, Singapore, and sub-Saharan Africa.

Final Thoughts

The 50 people on this list share one important characteristic: they are not waiting for AI in healthcare to be solved before they contribute to the conversation. They are working in the present, often with imperfect tools, incomplete evidence, and significant uncertainty, because the stakes are too high and the pace too fast to defer engagement until certainty arrives. Healthcare leaders who want to make good decisions about AI need to follow people who are doing the same.

The most useful thing you can do after reading this list is to start following three to five of these voices directly on LinkedIn or through their publications, and to engage with their arguments, ask questions, and bring their perspectives into the conversations happening in your own organisation. The global health AI conversation is not happening in a single place or being led by a single voice. It is distributed across research labs, health systems, regulatory bodies, community health centres, and advocacy organisations, in dozens of countries, in multiple languages, across multiple disciplines. The people on this list are some of its most important contributors.

For more on leadership in the healthcare sector, explore '50 Best Leadership Speakers for Hospitals (2026)' at https://www.consultclarity.org/post/50-best-leadership-speakers-for-hospitals-2026.

Jonno White helps healthcare leadership teams have the conversations they need to have about AI, alignment, and change. Whether you need a keynote for a health conference, a working session for your executive team, or a multi-day offsite to build clarity and direction, Jonno delivers practical tools and facilitation that your team will use on Monday morning. Email jonno@consultclarity.org. International travel is often far more affordable than organisations expect.

About the Author

Jonno White is a Certified Working Genius Facilitator, bestselling author, and leadership consultant who has worked with schools, corporates, and nonprofits across the UK, India, Australia, Canada, Mongolia, New Zealand, Romania, Singapore, South Africa, USA, Finland, Namibia, and more. His book Step Up or Step Out has sold over 10,000 copies globally, and his podcast The Leadership Conversations has featured 230+ episodes reaching listeners in 150+ countries. Jonno founded The 7 Questions Movement with 6,000+ participating leaders and achieved a 93.75% satisfaction rating for his Working Genius masterclass at the ASBA 2025 National Conference. Based in Brisbane, Australia, Jonno works globally and regularly travels for speaking and facilitation engagements. Organisations consistently find that international travel is far more affordable than expected.

To book Jonno for your next keynote, workshop, or facilitation session, email jonno@consultclarity.org.

Next Read: 50 Best Leadership Speakers for Hospitals (2026)

If you are searching for the best leadership speakers for hospital events, staff development days, or executive leadership programmes, you have probably noticed that the market divides roughly into two groups: global keynote names who command large fees and bring impressive credentials but limited healthcare specificity, and healthcare specialists who understand the sector deeply but have narrower platform and speaking profiles. The guide below profiles 50 speakers who cover every dimension of hospital leadership.

Keep reading: https://www.consultclarity.org/post/50-best-leadership-speakers-for-hospitals-2026