From Chaos to Control: Dr. Maureen Canellas on AI, Emergency Medicine & Why Most “AI Companies” Fake It

Artificial intelligence has moved from experimental promise to boardroom priority, but healthcare leaders are still trying to understand what is real, what is risky and what is just clever marketing. In emergency medicine, where patient status changes minute by minute and operational bottlenecks can ripple across an entire hospital, the stakes are especially high. The core challenge is not simply whether hospitals should adopt AI, but whether they can evaluate it critically enough to avoid wasting time, capital and clinician trust.So how should hospital executives, emergency physicians and operators decide which AI tools are worth trusting — and which ones are just hype?On this episode of I Don’t Care, host Dr. Kevin Stevenson speaks with Dr. Maureen Canellas, Associate Chief Medical Officer at UMass Memorial Medical Center, about the practical reality of AI in healthcare. Their conversation moves from emergency department operations and hospital governance to vendor skepticism, boarding, data standards, baseball and even NASA-inspired lessons in resource-constrained decision-making.Top insights from the talk…Most “AI” claims deserve scrutiny. Dr. Canellas argues that hospitals should look past slide decks and ask vendors detailed questions about data inputs, algorithm design, model behavior, privacy and whether the tool actually works in real time.Emergency departments need AI that respects clinical reality. In the ED, conditions change minute by minute. Dr. Canellas emphasizes that “real time” should mean real time, especially when vitals, labs and disposition decisions are constantly shifting.AI should help solve operational pain, not chase buzzwords. From boarding to throughput, Dr. Canellas says the best use cases are grounded in specific hospital problems, financial modeling and measurable return on investment — not vague promises to transform care.Dr. Maureen Canellas, MD, MBA, FACEP, is the Associate Chief Medical Officer at UMass Memorial Medical Center, an Assistant Professor of Emergency Medicine at UMass Chan Medical School and an emergency physician at a high-volume Level I trauma center. She is an MIT research affiliate in healthcare AI and uses operations research, machine learning and emergency department benchmarking to improve patient flow, care delivery and patient experience. Dr. Canellas also serves on the Emergency Department Benchmarking Alliance Board of Directors, chairs its education committee and previously trained in emergency medicine at the University of Chicago, where she was chief resident and led multiple operations and administration research projects.