AI Can’t Hear Your Stethoscope: Why Clinical Judgment Still Saves Lives | Dr. Sal D’Allura

AI can review research across many journals in seconds. But it can’t hear decreased breath sounds, notice that a patient can’t sit still, or sense that something is wrong when the vital signs look normal. In Part 2 of our conversation, Dr. Sal D’Allura, a supervising physician at Atrium Health Wake Forest Baptist with 36 years of clinical experience, moves from the classroom to the clinic. He shares the story of a young woman who came into urgent care late at night, sure she had pulled a muscle. Something about her presentation didn’t feel right. He sent her to the emergency room, where she was diagnosed with a large pulmonary embolism, and that decision saved her life. Dr. Sal explains why AI thinks in black and white while medicine lives in the gray, why AI-generated notes should never be “signed but not read,” and why the medical decision-making section must reflect what the clinician was actually thinking. He also shares where AI truly helps: saving time on documentation, summarizing research, supporting billing and administration, and giving clinicians face-to-face time with patients through ambient AI scribes. “The internet is black and white. Medicine is gray. That’s why it’s called an art.” In this episode, we cover: Why AI thinks in black and white, while medicine lives in the gray Ethics and end-of-life decisions AI can’t make A day as a supervising physician for PAs and NPs How he coaches clinicians on when a patient truly needs the ER Morbidity and mortality reviews then, and the lack of AI oversight now Why the medical decision-making section must reflect what the clinician was thinking The danger of AI notes that are “signed but not read” The pulmonary embolism case that clinical instinct caught Where AI helps: documentation, research, billing, and administration Ambient AI scribes and the return of face-to-face care Why he left private practice, and why “bigger is better” in today’s healthcare business AI anatomy and virtual dissection vs. the real operating room Should AI use in training be restricted by experience level? His final advice: treat AI as a resource, not a source Missed Part 1? Watch “AI in the Right Hands: Why Doctors Must Learn Medicine Before Using AI.” 📧 Connect with IDAAHub Follow us on: LinkedIn: https://www.linkedin.com/company/idaahub/ https://www.youtube.com/@IDAAHUB https://open.spotify.com/show/3V8Vuhqkibej5fUwtwkUMx https://podcasts.apple.com/us/podcast/the-idaa-hub-podcast-ai-in-finance-healthcare/id1848710327 📧 Connect with Host Host Deepti Kalghatgi : https://www.linkedin.com/in/deepti-kalghatgi/ 🌐 Visit: https://idaahub.com