AI Synthetic Patient — Demonstration Video

No. The platform supports the development of multiple synthetic patient profiles within a single program. Patient personas can be tailored to reflect different demographics, disease states, comorbidities, health literacy levels, and psychosocial or socioeconomic considerations, depending on the educational goals of the activity.

For this demonstration, the AI synthetic patient, “Danny,” is used in connection with an accredited educational activity focused on HIV diagnosis and communication. Learners first complete the accredited activity, which includes a faculty-led demonstration with a standardized patient. Following completion, learners may optionally engage with the AI synthetic patient to practice the communication skills and approaches that were demonstrated. This experience supports the application of learning rather than replacing formal instruction.

More broadly, AI synthetic patient experiences may be customized based on educational needs and goals and may be deployed with or without an accredited activity. “Danny” represents one example of how this technology can be used.

The AI synthetic patient is a non-accredited, supplemental tool used as part of a broader accredited CME/CE initiative. As an accredited provider, Talem Health adheres to all applicable ACCME and Joint Accreditation standards. The AI experience is clearly separated from the accredited content, free of commercial bias, and designed to support practice-based learning aligned with the educational objectives of the activity.

No. The AI synthetic patient is not a clinical decision-support tool. It is designed to simulate natural patient dialogue and behavior and does not provide medical advice, diagnoses, or treatment recommendations.

Yes. Participation in the synthetic patient experience is optional. Before engaging, learners review the purpose of the activity, the nature of the interaction, and the use of transcription and analysis for educational and outcomes purposes and must provide informed consent.

Each learner’s conversation generates structured data that supports outcomes assessment and educational improvement, including:

• Conversation transcripts capturing the full clinician learner–synthetic patient dialogue
• Perception and communication analysis, such as empathy, non-stigmatizing language, and non-biased body language
• Thematic insights that help identify common strengths, gaps, and learning patterns across participants will be analyzed in aggregate and used to inform outcomes reporting, educational refinement, and future program design