Two randomized experiments evaluate AI avatars and gender effects on dental care compliance and understanding.
Background Postcare instructions play a vital role in preventive health care. Traditionally, information on post–dental procedure care is provided through printed leaflets or brief verbal explanations, which often fail to engage patients or ensure adherence. In contrast, emerging artificial intelligence (AI) systems designed to emulate human interaction represent a promising yet understudied approach to improving patient communication and understanding. Objective This study evaluated the effectiveness of AI-generated avatars in communicating postprocedure dental instructions and determined how video characteristics, including avatar type (humanized, animated, or real person), AI identity disclosure, and avatar gender, impact intention to comply and comprehension. Methods Two preregistered online randomized experiments were conducted. Experiment 1 (n=650) compared the effectiveness of a real person with humanized and animated AI avatars, with and without disclosure of the AI’s identity. Experiment 2 (n=256) tested the effects of avatar gender on patient intention to comply. The primary outcomes were self-reported likelihood of following instructions, returning to the dentist, and comprehension. Results In experiment 1, humanized AI avatars were as effective as a real person in promoting intention to comply with instructions, while animated avatars significantly reduced this intention. AI disclosure did not affect intention to comply but reduced the reported likelihood of returning to the same dentist. In experiment 2, women reported a higher intention to comply with the instructions in the man avatar condition than in the woman avatar condition. Comprehension remained high and was unaffected across conditions. Conclusions Humanized AI avatars are an effective tool for delivering postprocedure instructions. However, design choices such as visual realism and avatar gender significantly impact patient trust and intention to comply. Avatar design should balance efficacy and ethics.
No takes yet. Share an insight, caveat, or question.
Benadof et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: