EACH Workshop: Automated Feedback in Clinical Communication Training: Opportunities and Pitfalls of Generative AI – Dr Thomas Kropmans

Thurs 28th May 2026 @4pm via Zoom

Please register your interest using this very quick form Irish EACH webinar registration form – Fill out form and Zoom details will be sent closer to the time.

High-quality feedback is essential for developing effective communication skills in clinical consultations. However, providing individualized and timely feedback at scale remains a challenge in medical education and recruitment settings such as Objective Structured Clinical Examinations (OSCEs), Multiple Mini Interviews (MMIs), and simulated patient consultations conducted both in-person and online.

Recent advances in generative artificial intelligence offer new opportunities to transform structured assessment data into automated, personalized feedback for learners. At the same time, AI-generated feedback must be carefully validated to ensure that it remains accurate, safe, and educationally meaningful.

This webinar explores how structured assessment rubrics and feedback templates can be combined with generative AI to produce automated feedback for communication skills training. The session draws on findings from a recent comparative research study evaluating AI-generated feedback from OpenAI GPT-4o, ChatGPT-5 and Antropic Claude Sonnet-4, which examined both preview feedback (for examiner verification) and portfolio feedback (student-facing) in OSCE settings. The study revealed important differences in accuracy, hallucination rates, tone of voice, and actionability, highlighting the need for governance and validation when deploying AI-driven feedback systems in high-stakes educational contexts.

Participants will explore practical strategies for designing feedback templates that translate assessment data into clear, constructive, and actionable feedback while minimizing risks associated with generative AI. The workshop will include examples from simulated patient consultations and provide guidance on evaluating AI-generated feedback before deployment in educational or recruitment environments.

Short Bio – Dr Thomas Kropmans
Dr Thomas Kropmans, retired senior lecturer Medical Informatics and Medical Education (UoG 2006-2022), is Co-Founder and Chief Executive Officer of Qpercom Ltd (founded in 2008), a company specializing in digital assessment platforms for healthcare education and recruitment. He has extensive experience in simulation-based assessment, OSCE design, and technology-enhanced feedback systems. Dr Kropmans has led multiple international projects focused on improving assessment quality and transparency in clinical education and recruitment processes, including digital scoring systems for OSCEs, MMIs, and video interviewing. His current research focuses on the application of generative AI for automated feedback generation, with particular emphasis on ensuring accuracy, safety, and actionable learning outcomes in high-stakes assessment environments.