Would you trust a doctor who uses ChatGPT during your visit? Today, we published a study in @npjDigitalMed. It’s the first survey of practicing clinicians on how they view peers who use generative AI in medical decision-making: nature.com/articles/s41746-0… The numbers are striking. In our randomized experiment of 276 clinicians, a physician who used GenAI as the primary tool was rated far lower in clinical skill (3.79 vs 5.93) and overall competence (3.71 vs 5.99) compared to one who did not use GenAI. That’s on a 7-point scale. Ratings of the care experience also dropped, from 4.48 to 3.08 stars. On a 5-star scale. Framing GenAI as a tool to verify decisions helped (clinical skill 4.99, competence 4.94) but didn’t close the gap. Independent of peer perceptions, clinicians saw the potential of using GenAI to improve healthcare: They believed GenAI improves accuracy (4.30) and viewed institution-customized GenAI even more favorably (4.96). Grateful to be working with a dream team, including Dr. Haiyang Yang @JHUCarey, Drs. Risa Wolf, Nestoras Mathioudakis, and Amy Knight, MD of @HopkinsMedicine, and Dr. @_yuna_ny (my former @JHUCarey @JohnsHopkinsSPH MBA/MPH student, now at @McKinsey). This work was supported by @JohnsHopkins Discovery Award. A huge thank you to @JHU_HBHI for a pilot grant that led to our Discovery Award. We also appreciate Drs. @jkvedar and @adamgdunn for such a constructive review process. Read the study (full-text open access): nature.com/articles/s41746-0…