PGY-AI: The Impact of Artificial Intelligence on Psychiatry Education-View Presentation

Video Transcription

Video Summary

A panel of psychiatry educators and clinicians explored how AI may shape medical education and clinical practice, emphasizing that the technology is advancing faster than the field can fully evaluate it.<br /><br />Speakers described current uses including AI-based tutoring and patient simulations, research support, predictive analytics, therapy chatbots, and documentation tools. AI scribes may reduce administrative burden, burnout, and time spent looking at screens, allowing clinicians to focus more on patients. However, relying on AI to generate assessments could weaken trainees’ clinical reasoning, documentation skills, and ability to develop an individual clinical voice. The panel discussed possible safeguards, such as limiting AI use to certain parts of notes or stages of training.<br /><br />Ethical and practical concerns included biased or culturally narrow outputs, fabricated information (“hallucinations”), data privacy and security, unclear liability, and the difficulty of explaining how AI reaches conclusions. Speakers also warned that overreliance could erode independent thinking and clinical intuition. If clinicians and AI disagree, responsibility may be unclear—especially if AI becomes a standard of care.<br /><br />The discussion stressed transparency and informed consent when AI is used in patient care, particularly in psychotherapy, where an AI listener may affect trust and the therapeutic relationship. Audience members shared examples of AI helping reveal changes in a patient’s thinking, while also raising concerns about vulnerable patients using chatbots during crises.<br /><br />The panel concluded that AI should be treated as a tool, not a substitute for clinical judgment. Responsible adoption will require AI literacy, ongoing evaluation, trainee involvement, attention to equity, and open collaboration between educators, clinicians, patients, and technology developers.

Keywords

psychiatry

artificial intelligence

medical education

clinical practice

AI scribes

predictive analytics

therapy chatbots

clinical reasoning

documentation skills

algorithmic bias

data privacy

informed consent

clinical judgment