A Doctor’s Warning: AI Still Can’t Replace Clinical Judgment
A New York Times opinion piece by a physician argues that artificial intelligence cannot do what doctors do, even as it becomes increasingly capable on narrow tasks. The essay lands in the middle of a broader debate over which parts of medicine are automatable and which depend on human judgment. For healthcare readers, the significance is not the argument itself, but how forcefully the profession is drawing a line around human responsibility.
The New York Times opinion piece reflects a growing counterweight to the exuberant claims surrounding medical AI. As models improve, the central debate is shifting from technical capability to professional identity and responsibility.
The physician’s argument is compelling because it frames medicine as more than pattern recognition. Real clinical work involves uncertainty, tradeoffs, patient values, communication, and ethical accountability. These elements are difficult to reduce to a model output, no matter how sophisticated the system becomes.
That perspective does not mean AI has no place in healthcare. On the contrary, many of the most valuable tools will likely support clinicians by reducing administrative burden, surfacing hidden patterns, or assisting with triage. But the op-ed suggests that the profession is increasingly unwilling to accept claims that AI can simply substitute for human judgment.
This matters because public expectations influence adoption. If patients begin to believe that AI is being used as a hidden replacement for doctors, trust may erode quickly. If clinicians believe the same thing, resistance will harden. Opinion pieces like this therefore shape the social boundary conditions around deployment.
Ultimately, the article captures a moment of recalibration. The healthcare sector is moving from “What can AI do?” to “What should remain distinctly human?” That question will define the next phase of the field.