Doctors Are Already Using AI to Triage the Hard Questions—And Patients May Never Know
The New York Times reports that physicians are increasingly turning to AI tools to help answer thorny medical questions, from diagnosis to treatment planning. The shift is happening quickly enough that the real issue is no longer whether AI is in the exam room, but how transparent that use will be.
AI is moving from a back-office curiosity to a quiet layer inside clinical decision-making. The New York Times piece captures a moment many health systems have been anticipating: doctors using AI to help navigate the most complex cases, often without making the tool’s role obvious to patients.
That matters because the value proposition of AI in medicine has always been two-sided. On one hand, it can compress information overload, summarize evidence, and help clinicians consider possibilities they might otherwise miss. On the other, it can blur accountability if patients assume a physician arrived at a recommendation entirely through personal expertise.
The article highlights a deeper shift in medical workflow. AI is not just replacing administrative tasks or drafting notes; it is increasingly being used as a cognitive aid. That makes governance harder, because the risk is not only bad output, but overreliance, hidden prompting, and a loss of visibility into when a machine shaped the final judgment.
The key question now is not whether AI should be used in clinical reasoning, but what standards should apply when it is. Health systems will need clearer norms for disclosure, documentation, and validation if they want patients to trust AI-assisted care rather than merely tolerate it.