The modern physician is no longer competing with disease alone. The new adversary is expectation: limitless, instantaneous, and increasingly shaped by algorithms.
In the span of a decade, the patient has evolved. First came “Dr. Google, MD” offering a chaotic but empowering flood of information. Then “Dr. PubMed, MD, PhD” sharpening the curious into quasi-specialists fluent in abstracts and endpoints. Now we face the almighty “Prof. AI”: tireless, confident, and available at 3 a.m., capable of generating differential diagnoses, treatment plans, and second opinions in seconds. For patients, this feels like liberation. For health care systems already stretched thin, it is an accelerant poured onto a slow-burning crisis.
The problem is not that patients are informed. On the contrary, an engaged, knowledgeable patient can elevate care, improve adherence, and foster genuine partnership. The problem is scale and mismatch. Health care systems across the world, strained by aging populations, workforce shortages, and finite resources, were never designed to meet infinite demand. Yet that is precisely what modern information ecosystems generate: the expectation of unlimited access, unlimited time, and unlimited solutions.
This is where the quiet wisdom attributed to Marcus Aurelius becomes unexpectedly relevant: “When you are disturbed by external things, it is not they that trouble you, but your own judgment of them.” Physicians today are not only overwhelmed by workload but by the psychological weight of expectation, of needing to validate, refute, or contextualize every AI-generated suggestion, every online diagnosis, every perceived omission. The disturbance is not merely the demand itself, but the implicit belief that all demands must be met.
Real physician voices, twice a week
Free, and one click to unsubscribe.
They cannot be. And pretending otherwise erodes both the clinician’s integrity and the system’s sustainability.
So how should we respond?
First, we must reassert the boundaries of medicine, not defensively, but transparently. Medicine is not an on-demand service industry; it is a prioritization system grounded in evidence, probability, and ethics. Not every test is indicated. Not every fear warrants intervention. Not every possibility deserves equal weight. Patients deserve explanations, not acquiescence.
Second, we must redefine the physician’s role in the age of AI. The clinician is no longer the primary source of information. That role is gone for good. Instead, the physician becomes an interpreter, a curator of relevance, and, critically, a steward of uncertainty. AI can generate answers; it cannot assume responsibility. It does not bear the consequences of overtesting, overtreatment, or misinterpretation. Physicians do.
Third, we must invest in communication as a clinical skill, not a soft accessory. The “demanding patient” is often not demanding for its own sake but responding to a landscape that amplifies fear and possibility in equal measure. Meeting that with dismissal fuels conflict; meeting it with structured dialogue transforms it. “I understand why this concerns you. Here’s what matters, and here’s what doesn’t, and why.” This is not just empathy; it is system preservation.
Finally, we must accept that dissatisfaction will not disappear. A system with finite capacity serving infinite expectations will always generate friction. The goal is not to eliminate that tension but to manage it without compromising care or burning out those who provide it.
The presence of “Prof. AI” does not signal the end of the physician. It clarifies their value. In a world overflowing with answers, what patients ultimately need is judgment, grounded, accountable, and human.
The challenge ahead is not to outcompete the machine, but to redefine what only we can provide, and to defend the limits that make that possible.
Paul Dranichnikov is a physician in Sweden.
