Teaching with Tech: How to Use Generative AI for Clinical Skills Development

Your students are likely already using AI (Artificial Intelligence), with recent research indicating a 26% rise in usage from 2024-25.  In this post, Grace Riley, Lecturer and Clinical Educator in the CEDAR Department at the University of Exeter, considers the ways in which Generative AI tools can be leveraged to support learners, and particularly those within a clinical programme.

Student use of AI is rising, with research estimating that 92% of university students were using AI in 2025, up from 66% in 2024 (Freeman, 2025). Moreover, whilst students argue that AI should be an integral part of their education experience (Jisc, 2023), only 36% of them receive institutional support to use AI (Freeman, 2025). Within clinical education, students are enhancing their traditional learning experience by incorporating AI tools like virtual patients and question-and-answer systems (Duan et al., 2025).

Generative AI (GenAI) can create realistic patient scenarios, ask questions, and provide feedback to the user, allowing students to practice interpersonal and communication skills and clinical decision making (Abd-Alrazaq et al., 2023). Utilising GenAI as a virtual patient offers an alternative to live roleplay by simulating realistic clinical interactions in a safe setting. In an era of digital healthcare, this could be particularly applicable to future practitioners required to deliver text-based support to patients. Below, I set out my three-step process to integrating GenAI tools into clinical training courses.

Step 1: Setup

I lecture on a clinical training course for NHS workers supporting patients with mental health difficulties by using CBT (Cognitive Behavioural Therapy) techniques. The competency to deliver interventions across all modes of support, including interactive text, is explicitly stated in the curriculum. At least half of the programme is dedicated to clinical skills development, affording the potential to use GenAI to support and enhance learning.

On the course, students are actively encouraged to use GenAI, engaging with GenAI roleplays to practise information gathering at assessment, and information giving within treatment. These are optional tasks, and the alternative is to complete a roleplay with a real person, to promote inclusion for those who were not as familiar or comfortable using AI tools.

When setting up these GenAI tasks for my learners, I provide a caveat with the instructions. GenAI cannot understand or feel emotions and lacks cultural competence (Wang et al., 2025). It may produce inaccurate or misleading information, and it cannot comprehend the meaning behind its responses. As GenAI tools typically filter out ‘harmful’ content, this limits the extent to which GenAI can be used to practice clinical skills such as risk assessment without ‘tricking’ the AI into bypassing these filters (Stapleton et al., 2023). Therefore, I emphasise that GenAI cannot and should not replace the experience of working with real patients. I also signpost to our internal student guidance on AI use to supplement their understanding and ensure they are using tools responsibly and ethically.

Depending on your students’ (and your) AI knowledge, you could also provide them with some suggestions of GenAI tools to use for this task. If you have scope, your students will likely find it helpful to be provided with training on how to responsibly access and use GenAI (Jisc, 2023).

Step 2: Provide prompts

You’ll need to provide a prompt for trainees to put into a GenAI tool to keep the conversation on track. Usually, the more specific the prompt is, the better. I use something like this for a roleplay task around information giving treatment:

“We will now roleplay a low intensity CBT treatment session via instant messenger. You will be a patient with panic disorder, and I am the practitioner. I will give you psychoeducation on panic symptoms so respond as if you are a patient.”

Here’s an example of my conversation with ChatGPT using the above prompt:

At the end of the roleplay, or at any point throughout, you can ask the AI to give you feedback on the messages, for example using the prompt:

“Can you now give me constructive feedback on my performance? For example, the clarity of my explanations, use of jargon, and interpersonal skills.”

Here’s a snippet of that exchange:

 

Step 3: Gather feedback

It’s pointless to make a change in teaching practice without exploring how effective it is. AdvanceHE has an excellent good practice guide to gathering student feedback, helping you to decide what method of feedback would be best for your cohort and task.

I used a Microsoft Form to gather written feedback around student experience and attitude towards using GenAI to support clinical skills development. Initial reports from students look promising:

• The AI sounds similar to a real patient
• The roleplay accurately simulated a live one with a real person
• The feedback given by the AI was fair and constructive
• It’s helpful to practise certain patient presentations by changing the initial prompt, e.g. adding “Sound anxious/reserved.”

GenAI won’t replace clinical experience gained from real patients and has some way to go regarding inclusivity and accuracy. But as a supplementary tool for clinical training, it has enormous potential. Why not give it a go within your clinical training?

Special acknowledgement to Josh Cable-May, Clinical Lead at Limbic, for the inspiration to use AI to develop trainee clinical skills development. 

References:

Abd-Alrazaq, A., AlSaad, R., Alhuwail, D., Ahmed, A., Healy, P. M., Latifi, S., … & Sheikh, J. (2023). Large language models in medical education: opportunities, challenges, and future directions. JMIR Medical Education, 9(1), e48291.

Duan, S., Liu, C., Rong, T., Zhao, Y., & Liu, B. (2025). Integrating AI in medical education: a comprehensive study of medical students’ attitudes, concerns, and behavioral intentions. BMC Medical Education, 25(1), 599.

Freeman, J. (2025). Student Generative AI Survey 2025. Higher Education Policy Institute. https://www.hepi.ac.uk/2025/02/26/student-generative-ai-survey-2025/#:~:text=In%202025%2C%20we%20find%20that,up%20from%2053%25%20in%202024.

Jisc. (2023). Student perceptions of generative AI. National centre for AI in tertiary education. National centre for AI in tertiary education: Student perceptions of generative AI report

Stapleton, L., Taylor, J., Fox, S., Wu, T., & Zhu, H. (2023). Seeing seeds beyond weeds: Green teaming generative ai for beneficial uses. arXiv preprint arXiv:2306.03097. https://doi.org/10.48550/arXiv.2306.03097

Wang, L., Bhanushali, T., Huang, Z., Yang, J., Badami, S., & Hightow-Weidman, L. (2025). Evaluating Generative AI in Mental Health: Systematic Review of Capabilities and Limitations. JMIR Mental Health, 12(1), e70014.

For more information please contact: