As artificial intelligence rapidly transforms medicine, healthcare leaders across the world are asking the same questions: How do we adopt AI responsibly? How do we preserve trust? How do we ensure technology strengthens rather than erodes the humanity of medicine?
Ironically, some of the answers may already exist — not in Silicon Valley, but in the foundational principles established decades ago at Bascom Palmer Eye Institute by one of the great leaders in academic medicine, Dr. Edward W.D. Norton.
I had the extraordinary privilege of training at Bascom Palmer during what many would consider its golden era, when all five legendary founders were still there: Dr. Edward Norton, Dr. Donald Gass, Dr. John Flynn, Dr. J. Lawton Smith, and Dr. Victor Curtin. These were not simply master clinicians and surgeons. They were institution builders who created a culture of excellence rooted in integrity, discipline, humility, intellectual rigor, and service.
One memory remains deeply etched in my mind. About a month before Dr. Norton passed away, I sat with him one afternoon in the rare book room at Bascom Palmer listening to him discuss his vision for the Institution. At the time, I viewed it largely through the lens of academic ophthalmology. Years later, I now realize he was describing something far broader: a framework for responsible leadership during times of profound transformation.
Today, as healthcare enters the AI era, the Norton Principles may be more relevant than ever.
The first principle was integrity. Dr. Norton believed integrity was non-negotiable. No matter how brilliant or accomplished someone might be, if they lacked integrity, they did not belong. That principle becomes critically important in artificial intelligence. AI systems are increasingly capable of influencing diagnoses, treatment decisions, workflows, documentation, research, and even patient interactions. But if the deployment of AI is driven primarily by profit, productivity metrics, or market pressure rather than patient welfare, the technology risks undermining the very trust upon which medicine depends.
Integrity in the AI era means transparency about limitations. It means acknowledging hallucinations, biases, data gaps, and uncertainty rather than overselling capability. It means resisting the temptation to use AI merely because it is available. Just because something can be automated does not necessarily mean it should be.
Another Norton Principle was credibility grounded in facts. Dr. Norton insisted that decisions be rooted in disciplined thinking, preparation, and evidence. He was never arbitrary. That lesson applies directly to AI adoption in healthcare today.
Much of the current AI conversation is fueled by hype cycles, venture capital enthusiasm, and fear of being left behind. Yet medicine has always required a higher evidentiary standard than other industries because human lives are involved. AI models may generate impressive outputs, but credibility requires validation, reproducibility, peer review, and rigorous clinical oversight. Physicians cannot abdicate responsibility to algorithms they do not understand. Fact-based confidence must remain the foundation of clinical decision-making.
Dr. Norton also believed in developing people rather than overshadowing them. He famously advised leaders to “let key faculty be the stars; don’t you try to be the brightest star in the universe.” That principle offers an important counterbalance to fears that AI will replace physicians.
READ: Sreedhar Potarazu | The Gods of Silicon Valley go to court: Altman, Musk, and the dangerous psychology behind AI and wealth (May 16, 2026)
The most effective future for AI in medicine is likely not replacement, but augmentation. AI should elevate human capability rather than diminish it. It should reduce administrative burdens, improve diagnostic support, expand access to care, and allow physicians to spend more time exercising the uniquely human aspects of medicine: empathy, judgment, contextual reasoning, ethical discernment, and compassion.
Dr. Norton also understood the importance of flexibility with strong personalities and unconventional thinkers. Bascom Palmer thrived because it cultivated innovation without sacrificing shared values. That lesson matters enormously today because AI development requires collaboration between clinicians, engineers, ethicists, regulators, data scientists, and entrepreneurs — groups that often approach problems from fundamentally different perspectives. Innovation ecosystems require tolerance for unconventional thinking while maintaining ethical guardrails.
Perhaps most importantly, Dr. Norton understood stewardship. He often described leadership as being a caretaker or gardener: “Pick the plants, cultivate the flowers, watch the blooms.” That concept may be the single best framework for thinking about AI governance in healthcare.
AI is not a finished product we simply purchase and deploy. It is an evolving ecosystem that requires careful cultivation. Leaders must continuously monitor outcomes, bias, safety, unintended consequences, and patient trust. Just as a gardener cannot abandon a growing garden after planting seeds, healthcare leaders cannot abdicate oversight after implementing AI systems. Stewardship requires constant vigilance.
Another overlooked Norton Principle was loyalty to the institution itself. There was an understanding at Bascom Palmer that individuals come and go, but the institution endures. Decisions therefore had to prioritize the long-term integrity of the institution over short-term personal gain.
That perspective is desperately needed in today’s AI landscape. Many organizations are rushing to implement AI tools to gain competitive advantage without fully understanding the long-term implications for patient privacy, physician autonomy, workforce stability, or public trust. Institutions must resist the temptation to prioritize short-term efficiency gains at the expense of enduring credibility. Once trust is lost in medicine, it is extraordinarily difficult to regain.
As I have moved through different seasons of life, I have come to appreciate these principles even more deeply. Life has a way of testing convictions. Ambition, adversity, distraction, ego, success, disappointment, and the sheer momentum of modern life can slowly pull any of us away from the frameworks that once grounded us. Often the drift is subtle, almost imperceptible.
But what makes enduring principles so powerful is that they remain there waiting for us. They become anchors to which we continually return.
As I reflect on the rapid acceleration of AI, I often think back to that afternoon conversation with Dr. Norton. He belonged to an era before machine learning, large language models, or generative AI. Yet the principles he championed may be precisely what we need to navigate this technological revolution responsibly.
Technology changes rapidly. Human nature does not.
Ambition, ego, financial incentives, competition, and the allure of innovation can easily pull individuals and institutions away from their foundational values. The challenge is not simply building powerful AI systems. The challenge is ensuring that our moral architecture evolves alongside our technological architecture.
The future of healthcare will undoubtedly involve artificial intelligence. That much is clear. But whether AI ultimately strengthens medicine or weakens it will depend less on the sophistication of the algorithms and more on the principles guiding the people who deploy them.
And perhaps that is why the Norton Principles remain so relevant today. They remind us that excellence in medicine has never been solely about technical capability. It has always been about character, stewardship, humility, integrity, and service to something larger than ourselves.
Those principles built one of the greatest ophthalmic institutions in the world. They may also help guide us safely into the age of artificial intelligence.
The Norton Principles — And Their Relevance to Artificial Intelligence in Healthcare
1. Integrity Above All Else
Integrity is non-negotiable. Character matters more than talent, credentials, or achievement.
In the AI era, integrity requires transparency about the limitations of algorithms, honesty about bias and hallucinations, and ensuring that patient welfare remains more important than productivity or profit.
2. Selflessness and Service to Community
Medicine is a calling rooted in compassion, humility, and obligation to society.
AI should be deployed to expand access to care, reduce disparities, alleviate physician burnout, and improve patient outcomes rather than merely optimize revenue or operational efficiency.
3. Credibility Through Facts and Preparation
Decisions should be evidence-based, disciplined, organized, and never arbitrary.
AI systems in healthcare must be rigorously validated, continuously monitored, and grounded in sound clinical evidence before widespread adoption.
4. Vision Coupled With Execution
Have a clear plan, balance priorities thoughtfully, and move decisively once prepared.
Healthcare organizations need thoughtful AI governance strategies with clear objectives, safeguards, physician oversight, and long-term planning rather than reactive implementation driven by hype.
5. Develop Others, Don’t Overshadow Them
Great leaders create environments where others can become the best versions of themselves.
AI should augment physicians, nurses, and healthcare teams — not diminish their autonomy, expertise, or humanity. Technology should elevate human potential rather than replace it.
6. Embrace Individuality and Intellectual Diversity
Innovation often comes from unconventional thinkers; cultivate excellence without demanding conformity.
Responsible AI development requires collaboration between clinicians, engineers, ethicists, regulators, entrepreneurs, and patients with diverse perspectives and expertise.
READ: Sreedhar Potarazu | The AI big bet: Laying off employees in exchange for chips and tokens (April 28, 2026)
7. With Respect
Strong leadership includes listening carefully, valuing differing opinions, and disagreeing without hostility.
Healthcare AI adoption must include open dialogue between stakeholders, particularly frontline clinicians and patients who will experience the real-world consequences of implementation.
8. Delegate Responsibility With Trust
Empower others with both responsibility and authority, while supporting their decisions.
AI systems should support informed clinical decision-making while preserving physician accountability and human judgment rather than creating blind dependence on automation.
9. Be a Steward and Caretaker
Leadership is like gardening: select carefully, nurture continuously, and cultivate long-term growth.
AI governance requires ongoing stewardship with continuous monitoring for bias, safety concerns, unintended consequences, and erosion of patient trust.
10. Loyalty to the Institution and Its Mission
Individuals are temporary; institutions endure. Protect and strengthen them for future generations.
Healthcare organizations must prioritize long-term public trust, ethical standards, and institutional credibility over short-term competitive advantage in AI deployment.
11. Invest in Excellence Rather Than Personal Gain
Build organizations of lasting quality by prioritizing mission, scholarship, and patient care over short-term rewards.
AI should be implemented to improve quality, safety, education, and patient care — not solely to maximize billing efficiency or reduce labor costs.
12. Remain Grounded in Humility
Technical brilliance without humility eventually erodes judgment, trust, and wisdom.
As AI systems become increasingly powerful, humility becomes even more essential. Physicians and technologists alike must recognize that intelligence alone does not equal wisdom, judgment, or moral clarity.
