The following blog is an excerpt from our white paper, The Human Algorithm: Confronting the Role of AI in Healthcare.
You may remember the phrase “show your work” from elementary school math lessons, but in medicine, it’s not just a suggestion; it’s a mandate.
When someone’s health is on the line, clinicians must go beyond a simple recommendation to provide the real “why” behind their decision. Patients (rightfully) want to know the reasoning for a new treatment or diagnostic test, and how it fits their specific needs.
This is where the use of AI for clinical purposes gets tricky. It is incredibly helpful in selecting and summarizing content across the internet, but its ability to show how it came to its conclusion is limited to simply linking to sources.
A clinician is treating a patient newly diagnosed with Type 2 Diabetes. They turn to an AI tool for guidance on first-line treatment options. In seconds, the AI aggregates evidence-based recommendations from trusted sources and offers a concise summary:
“Metformin is generally considered the first-line therapy for Type 2 diabetes. Here are links to ADA guidelines and recent clinical studies supporting this approach.”
This is undeniably useful. It saves time, pulls relevant evidence, and offers a solid starting point. But it stops short of what real clinical care requires.
AI can’t tell you:
Whether this patient’s kidney function makes metformin appropriate
How their past medication experiences influence adherence
Whether lifestyle factors or co-morbidities warrant an alternative
Or how shared decision-making might shape the choice
AI offers information. Only clinicians can offer interpretation.
AI may recommend an option based on patterns in the data — but it cannot meaningfully articulate the reasoning or risk-benefit analysis behind it. Clinicians must still evaluate the proposal, weigh its appropriateness, and justify the path forward.
In the end, AI can surface recommendations. It can accelerate the research. It can streamline the workload. But it’s the human layer — critical thinking, professional judgment, and clinical context that transforms raw data into trustworthy care.
To further explore how AI can enhance clinical judgment and get exclusive insights from the Healthcasts community, explore the complete findings.
Looking for more insights? Follow Healthcasts on LinkedIn, Facebook, and Instagram, then join our community to connect with a community of verified peers today.