From good prompts to great outcomes: practical AI skills for clinical practice

Most clinicians who try AI walk away unimpressed, and it usually isn't the tool's fault. It's the prompt. This session starts from the ground up with a plain-language explanation of what AI actually is and how these tools work, so anyone who missed part 1 can follow from the first slide. From there we build a simple, repeatable framework for writing prompts that produce genuinely useful clinical output instead of generic filler.

The majority of the session is hands-on, with prompts built, broken and rebuilt in real time: turning messy notes into a summary worth sharing with a team, and building a clear explanation that is then adapted for a different audience and a different reading level. We will also cover the skills that matter most for safe use: letting AI write and improve its own prompts, grounding answers in real documents so the tool cannot invent things, spotting hallucinations, and recognising when AI is simply the wrong tool. Every demonstration is run without any patient information, and the reasoning behind that is part of the teaching. GDPR, confidentiality and professional judgement run throughout.