Uploaded July 2026 | Updated September 2026, 2 weeks ago
Is the AI tool you're using biased against your patients?
AI tools are being deployed across emergency departments, cardiology units, radiology suites, and primary care clinics worldwide. But a growing body of evidence shows that many of these tools perform significantly worse for women, Black patients, older adults, and those from lower socioeconomic backgrounds.
As a clinician, you are the last line of defense before a biased AI recommendation reaches your patient.
In this webinar, you'll learn:
1. Where AI bias originates—across data collection, algorithmic optimization, clinical implementation, and generative AI
2. How to audit any AI tool using practical frameworks including Aequitas, PROBAST, and AIF360
3. What questions to ask a vendor before your hospital deploys a new AI system
4. How to stay clinically accountable when AI and your judgment disagree
Speaker:
DJ Apakama, MD MS FACEP, is an Assistant Professor in the Windreich Department of Artificial Intelligence and Human Health and the Department of Emergency Medicine at the Icahn School of Medicine at Mount Sinai. His research focuses on algorithmic bias, health equity, and the responsible integration of AI in clinical settings. He is a practicing emergency physician and a leading voice on clinical AI stewardship.
0:00 Introduction and welcome
5:28 Talk begins: the AI bias diagnostic framework
14:08 Dimension 1: Data bias
23:05 Dimension 2: Algorithm bias
33:00 Dimension 3: Implementation bias
41:17 The LLM amplification effect
53:10 Practical tools and five questions to ask before deploying AI
Is the AI tool you're using biased against your patients?
AI tools are being deployed across emergency departments, cardiology units, radiology suites, and primary care clinics worldwide. But a growing body of evidence shows that many of these tools perform significantly worse for women, Black patients, older adults, and those from lower socioeconomic backgrounds.
As a clinician, you are the last line of defense before a biased AI recommendation reaches your patient.
In this webinar, you'll learn:
1. Where AI bias originates—across data collection, algorithmic optimization, clinical implementation, and generative AI
2. How to audit any AI tool using practical frameworks including Aequitas, PROBAST, and AIF360
3. What questions to ask a vendor before your hospital deploys a new AI system
4. How to stay clinically accountable when AI and your judgment disagree
Speaker:
DJ Apakama, MD MS FACEP, is an Assistant Professor in the Windreich Department of Artificial Intelligence and Human Health and the Department of Emergency Medicine at the Icahn School of Medicine at Mount Sinai. His research focuses on algorithmic bias, health equity, and the responsible integration of AI in clinical settings. He is a practicing emergency physician and a leading voice on clinical AI stewardship.
0:00 Introduction and welcome
5:28 Talk begins: the AI bias diagnostic framework
14:08 Dimension 1: Data bias
23:05 Dimension 2: Algorithm bias
33:00 Dimension 3: Implementation bias
41:17 The LLM amplification effect
53:10 Practical tools and five questions to ask before deploying AI
